A comprehensive monitoring and early warning method and system for mine disasters based on microseismic monitoring technology

Abstract

The present invention relates to the field of mine disaster monitoring technology, and specifically to a mine disaster comprehensive monitoring and early warning method and system based on microseismic monitoring technology, including the following steps: collecting surface microseismic wave data and underground microseismic wave data in a mine area; calculating the source position and source energy through underground microseismic wave data; migrating the mapping relationship between underground microseismic wave data and source position and source energy into the mapping relationship between surface microseismic wave data and source position and source capacity; encapsulating the mapping relationship between surface microseismic wave data and source position and source capacity into a neural network to obtain a mine disaster monitoring model. The present invention can achieve disaster monitoring and early warning by determining the source position and source energy from surface microseismic wave data by constructing a mine disaster monitoring model, calculating the source information by collecting seismic wave signals on the ground, and laying out monitoring devices on the ground, which is simple to operate and ensures monitoring safety.

Description

Technical Field

[0001] The present invention relates to the technical field of mine disaster monitoring, and in particular to a mine disaster comprehensive monitoring and early warning method and system based on microseismic monitoring technology. Background Art

[0002] The basic principle of microseismic monitoring technology is to collect and acquire seismic wave signals emitted by rock destruction or rock fracture through sensors. The seismic wave signals are processed and analyzed to obtain information such as the location, magnitude, energy, and seismic moment of the mine earthquake. Therefore, the use of microseismic monitoring technology can realize comprehensive monitoring of mine disasters.

[0003] When the existing technology uses microseismic monitoring technology to conduct comprehensive monitoring of mine disasters, it is usually adopted to collect seismic wave signals underground to calculate the earthquake source information, which ensures the accuracy of the calculation of the earthquake source information. However, the underground layout of the equipment for collecting seismic wave signals is highly complex and has great terrain limitations. In the process of underground equipment layout, it may even cause rock damage or rock fracture, thereby destroying the safety of the rock mass. Therefore, the calculation of earthquake source information by collecting seismic wave signals underground is complicated to operate and has safety defects. Summary of the invention

[0004] The purpose of the present invention is to provide a comprehensive monitoring and early warning method and system for mine disasters based on microseismic monitoring technology, so as to solve the technical problems in the prior art of calculating source information by collecting seismic wave signals underground, which is complicated to operate and has safety defects.

[0005] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0006] A comprehensive monitoring and early warning method for mine disasters based on microseismic monitoring technology comprises the following steps:

[0007] Collect surface microseismic wave data and underground microseismic wave data in the mining area;

[0008] The earthquake source location and energy are calculated through underground microseismic wave data;

[0009] The mapping relationship between underground microseismic wave data and source location and source energy is transferred to the mapping relationship between surface microseismic wave data and source location and source energy;

[0010] The mapping relationship between surface microseismic wave data and source location and source capacity is encapsulated into a neural network to obtain a mine disaster monitoring model for determining the source location and source energy based on the surface microseismic wave data.

[0011] As a preferred solution of the present invention, the method for calculating the earthquake source position through underground microseismic wave data includes:

[0012] The dynamic equation of microseismic wave propagation is established using a spherical coordinate system. The dynamic equation of microseismic wave propagation is: in, In the formula, r is the distance between the earthquake source and the underground microseismic wave detector, t is the propagation time of the microseismic wave, and Y r (t) is the microseismic wave data collected by the underground microseismic wave detector at the tth microseismic wave propagation time, Y 0 (tr / v) is the microseismic wave data of the earthquake source at the tr / vth microseismic wave propagation time, and v is the propagation speed of the microseismic wave;

[0013] The dynamic equation of microseismic wave propagation is transformed using the earthquake source tomography positioning function to obtain the earthquake source function, which is: 0 (t) = k(r)*A p (t), where Where Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, r j is the distance from the earthquake source to the jth underground microseismic detector, A p (t) is the maximum capacity value in the earthquake source tomography positioning function, k(r) is the energy diffusion compensation factor, and M is the total number of underground microseismic geophones.

[0014] As a preferred solution of the present invention, the method for calculating the source energy through underground microseismic wave data includes:

[0015] The kinematic equation of energy flux density is constructed, and the kinematic equation of energy flux density is: In the formula, I r (t) is the energy flux density collected by the underground microseismic wave detector at the tth microseismic wave propagation time, ρ is the propagation medium of the microseismic wave, v is the propagation speed of the microseismic wave, u is the shear modulus, r is the distance between the earthquake source and the underground microseismic wave detector, and t is the propagation time of the microseismic wave;

[0016] Using T = (tr / v) Convert and get

[0017] Depend on The source energy is obtained by performing an integral operation, and the source energy is: Where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment.

[0018] As a preferred solution of the present invention, a method for migrating the mapping relationship between underground microseismic wave data and source location and source energy into the mapping relationship between surface microseismic wave data and source location and source energy includes:

[0019] Establish a mapping relationship between underground microseismic wave data, source location and source energy. The mapping relationship between underground microseismic wave data, source location and source energy is {E(t), Y 0 (t)}=F({Y r,j (t)|j∈[1,M]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment, Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, Y r,j (t) is the microseismic wave data collected by the jth underground microseismic wave detector at the tth microseismic wave propagation time, F is the mapping function of underground microseismic wave data, source position and source energy, and M is the total number of underground microseismic wave detectors;

[0020] The mapping relationship between underground microseismic wave data, source location and source energy is replaced by surface microseismic wave data to obtain the mapping relationship between surface microseismic wave data, source location and source energy. The mapping relationship between surface microseismic wave data, source location and source energy is {E(t), Y 0 (t)}=H({X r,i (t)|i∈[1,N]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment, Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, X r,i (t) is the microseismic wave data collected by the i-th surface microseismic wave detector at the t-th microseismic wave propagation time, H is the mapping function of the surface microseismic wave data with the source position and source energy, and N is the total number of surface microseismic wave detectors.

[0021] As a preferred solution of the present invention, the construction of the mine disaster monitoring model includes:

[0022] {X r,i (t)|i∈[1,N]} as the input of BP neural network, and E(t), Y 0 (t) as the output item of the BP neural network, the BP neural network is used to train the input item and the output item, encapsulate the mapping relationship between the surface microseismic wave data and the source position and source capacity, and obtain the mine disaster monitoring model;

[0023] The mine disaster monitoring model is {E(t), Y 0 (t)}=BP({X r,i (t)|i∈[1,N]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment, Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation moment, X r,i(t) is the microseismic wave data collected by the i-th surface microseismic wave detector at the t-th microseismic wave propagation time, BP is the BP neural network, and N is the total number of surface microseismic wave detectors.

[0024] As a preferred solution of the present invention, the loss function of the mine disaster monitoring model is: Loss = λ 1 ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 +λ 2 ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 ;

[0025] In the formula, Loss is the loss function, λ 1 and λ 2 They are respectively ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 The weight of {E(t), Y 0 (t)} BP is the output of the mine disaster monitoring model, E(t), Y 0 (t), {E(t), Y 0 (t)} F is E(t) calculated from underground microseismic wave data, Y 0 (t), {E(t), Y 0 (t)} real is E(t), Y 0 The true value of (t), ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 All are L1 norm forms.

[0026] As a preferred solution of the present invention, the weight λ 1 and λ 2 They are:

[0027] λ 1 = -(1-||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 )log(||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 );

[0028] λ 2 = -(1-||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 )log(||{E(t),Y 0 (t)} BP -{E(t),

[0029] Y 0 (t)} real || 1 );

[0030] In the formula, λ 1 and λ 2 They are respectively ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 The weight of {E(t), Y 0 (t)} BP E(t), Y is the output of the mine disaster monitoring model 0 (t), {E(t), Y 0 (t)} F is E(t) calculated from underground microseismic wave data, Y 0 (t), {E(t), Y 0 (t)}real is E(t), Y 0 The true value of (t), ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 All are L1 norm forms.

[0031] As a preferred solution of the present invention, the surface microseismic wave data and the underground microseismic wave data are normalized.

[0032] As a preferred solution of the present invention, the Y output by the mine disaster monitoring model is 0 (t) Compare with the preset radiation energy threshold for disaster warning,

[0033] When the mine disaster monitoring model outputs Y 0 (t) exceeds the preset radiation energy threshold for disaster warning, the corresponding mine disaster monitoring model output E(t) is marked as the disaster source and an early warning is issued;

[0034] When the mine disaster monitoring model outputs Y 0 If (t) does not exceed the preset radiation energy threshold for disaster warning, the E(t) output by the corresponding mine disaster monitoring model will be marked as a non-disaster source.

[0035] As a preferred solution of the present invention, the present invention provides a comprehensive detection and early warning method for mine disasters based on microseismic monitoring technology. According to the comprehensive monitoring and early warning method for mine disasters based on microseismic monitoring technology, the system includes:

[0036] A data acquisition unit, used to collect surface microseismic wave data and underground microseismic wave data in the mining area;

[0037] A data preprocessing unit, used to calculate the earthquake source position and earthquake source energy through underground microseismic wave data;

[0038] and for migrating the mapping relationship between underground microseismic wave data and source location and source energy into the mapping relationship between surface microseismic wave data and source location and source energy;

[0039] A deep learning unit is used to encapsulate the mapping relationship between surface microseismic wave data and the source location and source capacity into a neural network to obtain a mine disaster monitoring model for determining the source location and source energy based on the surface microseismic wave data;

[0040] The monitoring and early warning unit is used to monitor and warn of mine disasters in the mining area using a mine disaster monitoring model.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] The present invention constructs a mine disaster monitoring model, which can determine the source location and source energy of the earthquake using surface microseismic wave data to complete disaster monitoring and early warning. The earthquake source information is calculated by collecting seismic wave signals on the ground, and the monitoring device is arranged on the ground. The operation is simple and the monitoring safety is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0044] Figure 1 A flow chart of a comprehensive monitoring and early warning method for mine disasters provided by an embodiment of the present invention;

[0045] Figure 2 A block diagram of a comprehensive monitoring and early warning system for mine disasters provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, the present invention provides a comprehensive monitoring and early warning method for mine disasters based on microseismic monitoring technology, comprising the following steps:

[0048] Collect surface microseismic wave data and underground microseismic wave data in the mining area;

[0049] The earthquake source location and energy are calculated through underground microseismic wave data;

[0050] The mapping relationship between underground microseismic wave data and source location and source energy is transferred to the mapping relationship between surface microseismic wave data and source location and source energy;

[0051] The mapping relationship between surface microseismic wave data and source location and source capacity is encapsulated into a neural network to obtain a mine disaster monitoring model for determining the source location and source energy based on the surface microseismic wave data.

[0052] The present invention constructs a mine disaster monitoring model, encapsulating the mapping relationship between surface microseismic wave data and the source location and source capacity, thereby realizing the determination of the source location and source energy based on the surface microseismic wave data, and then analyzing, judging and warning of disasters based on the source location and source energy.

[0053] The present invention can determine the source location and source energy based on surface microseismic wave data, and can deploy microseismic monitoring on the surface of a mining area without the need to install equipment underground, thereby avoiding rock damage or rock fractures that may damage rock safety during underground equipment layout, thereby changing complex underground monitoring into simple and safe above-ground monitoring.

[0054] The monitoring device for surface microseismic wave data is arranged on the surface, and the layout is simple. However, the accuracy of calculating the source location and source energy is not as good as that of underground microseismic wave monitoring. However, the layout is simpler and safer than that of underground microseismic wave monitoring.

[0055] The data used for training in the mine disaster monitoring model of the present invention (i.e., the earthquake source location and earthquake source energy) comes from the measurement of underground microseismic wave data. Therefore, the mine disaster monitoring model can obtain the measurement results of the earthquake source location and earthquake source energy of underground microseismic wave data based on the surface microseismic wave data, so that the originally simple but inaccurate surface monitoring can obtain the measurement results of underground monitoring with high accuracy, that is, through simple surface monitoring, high-accuracy earthquake source location and earthquake source energy measurement results are obtained (low-precision monitoring obtains high-precision earthquake source calculation results), thereby improving the model performance of the mine disaster monitoring model.

[0056] In the training of the mine disaster monitoring model, the present invention uses ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 is the loss function, where ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 Measure, E(t), Y output by mine disaster monitoring model0 (t) and E(t), Y calculated by underground microseismic wave data 0 (t), the smaller the value, the more the mine disaster monitoring model imitates the calculation process of underground microseismic wave data, so as to obtain the calculation results of the source position and source energy of the underground microseismic wave data according to the surface microseismic wave data, that is, the surface microseismic data obtains high-precision source calculation results (the surface monitoring accuracy is equivalent to the underground monitoring accuracy), || {E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 E(t), Y, which measures the output of the mine disaster monitoring model 0 (t) and E(t), Y 0 The smaller the value is, the more the mine disaster monitoring model imitates the real measurement process, so as to obtain the real earthquake source position and earthquake source energy measurement results based on the surface microseismic wave data, that is, the surface microseismic data further obtains high-precision earthquake source measurement results (the surface monitoring accuracy is equivalent to the real monitoring accuracy).

[0057] At the same time, in ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 The weight λ is added 1 and λ 2 , so that when the mine disaster monitoring model outputs E(t), Y 0 (t) and E(t), Y calculated by underground microseismic wave data 0 When the difference between (t) is greater, the surface monitoring accuracy cannot be equivalent to underground monitoring, and its weight λ 1 will be adaptively improved, thus shifting the focus of model training to imitating the underground monitoring and measurement process. Similarly, in the output of the mine disaster monitoring model, E(t), Y 0 (t) and E(t), Y 0 The greater the difference between the actual value of (t), the greater the weight λ 2 will be adaptively improved, so that the focus of model training will be tilted to imitate the real monitoring and measurement process, so that the weight λ 1 and λ 2It can help train the model so that the final output result is close to the underground result and close to the true value, that is, the model is equivalent to underground monitoring, and a high-performance model with high-precision monitoring results can be obtained.

[0058] Methods for calculating the earthquake source location using underground microseismic wave data include:

[0059] The dynamic equation of microseismic wave propagation is established using the spherical coordinate system. The dynamic equation of microseismic wave propagation is: in, In the formula, r is the distance between the earthquake source and the underground microseismic wave detector, t is the propagation time of the microseismic wave, and Y r (t) is the microseismic wave data collected by the underground microseismic wave detector at the tth microseismic wave propagation time, Y 0 (tr / v) is the microseismic wave data of the earthquake source at the tr / vth microseismic wave propagation time, and v is the propagation speed of the microseismic wave;

[0060] The earthquake source tomography location function is: A p (t) = ∫Y r (t+p)dr, where r j = dis[(x j ,y j ,z j ),(x 0 ,y 0 ,z 0 )], where dis is the Euclidean distance function, (x j ,y j ,z j ) is the position coordinate of the jth underground microseismic detector, (x 0 ,y 0 ,z 0 ) are the location coordinates of the earthquake source;

[0061] A p (t) = ∫Y r (t+p)dr and Convert and get

[0062]

[0063] According to the discrete distribution of underground microseismic geophones, Converted into discrete expression, we get

[0064] The dynamic equation of microseismic wave propagation is transformed using the source tomography positioning function to obtain the source function, which is: Y 0 (t) = k(r)*A p (t), where Where Y0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, r j is the distance from the earthquake source to the jth underground microseismic detector, A p (t) is the maximum capacity value in the earthquake source tomography positioning function, k(r) is the energy diffusion compensation factor, and M is the total number of underground microseismic geophones;

[0065] Methods for calculating source energy from underground microseismic wave data include:

[0066] The kinematic equation of energy flux density is constructed. The kinematic equation of energy flux density is: In the formula, I r (t) is the energy flux density collected by the underground microseismic wave detector at the tth microseismic wave propagation time, ρ is the propagation medium of the microseismic wave, v is the propagation speed of the microseismic wave, u is the shear modulus, r is the distance between the earthquake source and the underground microseismic wave detector, and t is the propagation time of the microseismic wave;

[0067] Using T = (tr / v) Convert and get

[0068] Depend on The source energy is obtained by integration operation. The source energy is: Where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment.

[0069] The method of transferring the mapping relationship between underground microseismic wave data and the location and energy of the earthquake source to the mapping relationship between surface microseismic wave data and the location and energy of the earthquake source includes:

[0070] The mapping relationship between underground microseismic wave data, source location and source energy is established. The mapping relationship between underground microseismic wave data, source location and source energy is {E(t), Y 0 (t)}=F({Y r,j (t)|j∈[1,M]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment, Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, Y r,j (t) is the microseismic wave data collected by the jth underground microseismic wave detector at the tth microseismic wave propagation time, F is the mapping function of underground microseismic wave data, source position and source energy, and M is the total number of underground microseismic wave detectors;

[0071] The mapping relationship between underground microseismic wave data, source location and source energy is replaced by surface microseismic wave data to obtain the mapping relationship between surface microseismic wave data, source location and source energy. The mapping relationship between surface microseismic wave data, source location and source energy is {E(t), Y 0 (t)}=H({X r,i (t)|i∈[1,N]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment, Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, X r,i (t) is the microseismic wave data collected by the i-th surface microseismic wave detector at the t-th microseismic wave propagation time, H is the mapping function of the surface microseismic wave data with the source position and source energy, and N is the total number of surface microseismic wave detectors.

[0072] The present invention constructs a mine disaster monitoring model, encapsulating the mapping relationship between surface microseismic wave data and the source location and source capacity, thereby realizing the determination of the source location and source energy based on the surface microseismic wave data, and then analyzing, judging and warning of disasters based on the source location and source energy.

[0073] The construction of mine disaster monitoring model includes:

[0074] {X r,i (t)|i∈[1,N]} as the input of BP neural network, and E(t), Y 0 (t) as the output item of the BP neural network, the BP neural network is used to train the input item and the output item, encapsulate the mapping relationship between the surface microseismic wave data and the source position and source capacity, and obtain the mine disaster monitoring model;

[0075] The mine disaster monitoring model is {E(t), Y 0 (t)}=BP({X r,i (t)|i∈[1,N]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment, Y 0 (t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, X r,i (t) is the microseismic wave data collected by the i-th surface microseismic wave detector at the t-th microseismic wave propagation time, BP is the BP neural network, and N is the total number of surface microseismic wave detectors.

[0076] The loss function of the mine disaster monitoring model is: Loss = λ 1 ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F ||1 +λ 2 ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 ;

[0077] In the formula, Loss is the loss function, λ 1 and λ 2 They are respectively ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 The weight of {E(t), Y 0 (t)} BP E(t), Y is the output of the mine disaster monitoring model 0 (t), {E(t), Y 0 (t)} F is E(t) calculated from underground microseismic wave data, Y 0 (t), {E(t), Y 0 (t)} real is E(t), Y 0 The true value of (t), ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 All are L1 norm forms.

[0078] Weight λ 1 and λ 2 They are:

[0079] λ 1 = -(1-||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 )log(||{E(t),Y0 (t)} BP -{E(t),Y 0 (t)} F || 1 );

[0080] λ 2 = -(1-||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 )log(||{E(t),Y 0 (t)} BP -{E(t),

[0081] Y 0 (t)} real || 1 );

[0082] In the formula, λ 1 and λ 2 They are respectively ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 The weight of {E(t), Y 0 (t)} BP is the output of the mine disaster monitoring model, E(t), Y 0 (t), {E(t), Y 0 (t)} F is E(t) calculated from underground microseismic wave data, Y 0 (t), {E(t), Y 0 (t)} real is E(t), Y 0 The true value of (t), ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 All are L1 norm forms.

[0083] In the training of the mine disaster monitoring model, the present invention uses ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 is the loss function, where ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 Measure, E(t), Y output by mine disaster monitoring model 0 (t) and E(t), Y calculated by underground microseismic wave data 0 (t), the smaller the value, the more the mine disaster monitoring model imitates the calculation process of underground microseismic wave data, so as to obtain the calculation results of the source position and source energy of the underground microseismic wave data according to the surface microseismic wave data, that is, the surface microseismic data obtains high-precision source calculation results (the surface monitoring accuracy is equivalent to the underground monitoring accuracy), || {E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 E(t), Y, which measures the output of the mine disaster monitoring model 0 (t) and E(t), Y 0 The smaller the value is, the more the mine disaster monitoring model imitates the real measurement process, so as to obtain the real earthquake source position and earthquake source energy measurement results based on the surface microseismic wave data, that is, the surface microseismic data further obtains high-precision earthquake source measurement results (the surface monitoring accuracy is equivalent to the real monitoring accuracy).

[0084] At the same time, in ||{E(t), Y 0 (t)} BP -{E(t),Y 0 (t)} F || 1 and ||{E(t),Y 0 (t)} BP -{E(t),Y 0 (t)} real || 1 The weight λ is added1 and λ 2 , so that when the mine disaster monitoring model outputs E(t), Y 0 (t) and E(t), Y calculated by underground microseismic wave data 0 When the difference between (t) is greater, the surface monitoring accuracy cannot be equivalent to underground monitoring, and its weight λ 1 will be adaptively improved, thus shifting the focus of model training to imitating the underground monitoring and measurement process. Similarly, in the output of the mine disaster monitoring model, E(t), Y 0 (t) and E(t), Y 0 The greater the difference between the actual value of (t), the greater the weight λ 2 will be adaptively improved, so that the focus of model training will be tilted to imitate the real monitoring and measurement process, so that the weight λ 1 and λ 2 It can help train the model so that the final output result is close to the underground result and close to the true value, that is, the model is equivalent to underground monitoring, and a high-performance model with high-precision monitoring results can be obtained.

[0085] The surface microseismic wave data and the underground microseismic wave data are normalized.

[0086] The output of the mine disaster monitoring model is Y 0 (t) Compare with the preset radiation energy threshold for disaster warning,

[0087] When the mine disaster monitoring model outputs Y 0 (t) exceeds the preset radiation energy threshold for disaster warning, the corresponding mine disaster monitoring model output E(t) is marked as the disaster source and an early warning is issued;

[0088] When the mine disaster monitoring model outputs Y 0 If (t) does not exceed the preset radiation energy threshold for disaster warning, the E(t) output by the corresponding mine disaster monitoring model will be marked as a non-disaster source.

[0089] like Figure 2 As shown, the present invention provides a comprehensive detection and early warning method for mine disasters based on microseismic monitoring technology. According to a comprehensive monitoring and early warning method for mine disasters based on microseismic monitoring technology, the system includes:

[0090] Data acquisition unit (microseismic wave detector), used to collect surface microseismic wave data and underground microseismic wave data in the mining area;

[0091] A data preprocessing unit, used to calculate the earthquake source position and earthquake source energy through underground microseismic wave data;

[0092] and for migrating the mapping relationship between underground microseismic wave data and source location and source energy into the mapping relationship between surface microseismic wave data and source location and source energy;

[0093] A deep learning unit is used to encapsulate the mapping relationship between surface microseismic wave data and the source location and source capacity into a neural network to obtain a mine disaster monitoring model for determining the source location and source energy based on the surface microseismic wave data;

[0094] The monitoring and early warning unit is used to monitor and warn of mine disasters in the mining area using a mine disaster monitoring model.

[0095] The present invention constructs a mine disaster monitoring model, which can determine the source location and source energy of the earthquake using surface microseismic wave data to complete disaster monitoring and early warning. The earthquake source information is calculated by collecting seismic wave signals on the ground, and the monitoring device is arranged on the ground. The operation is simple and the monitoring safety is guaranteed.

[0096] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A comprehensive monitoring and early warning method for mine disasters based on microseismic monitoring technology, characterized in that: The following steps are involved: Collect surface microseismic wave data and underground microseismic wave data in the mining area; The earthquake source location and energy are calculated through underground microseismic wave data; The mapping relationship between underground microseismic wave data and the location and energy of the earthquake source is transferred to the mapping relationship between surface microseismic wave data and the location and energy of the earthquake source; The mapping relationship between surface microseismic wave data and the earthquake source location and earthquake source energy is encapsulated into a neural network to obtain a mine disaster monitoring model for determining the earthquake source location and earthquake source energy based on the surface microseismic wave data; The method of transferring the mapping relationship between underground microseismic wave data and the earthquake source position and the earthquake source energy into the mapping relationship between surface microseismic wave data and the earthquake source position and the earthquake source energy includes: The mapping relationship between underground microseismic wave data, source location and source energy is established. The mapping relationship between underground microseismic wave data, source location and source energy is {E(t), Y0(t)}=F({Y r,j (t)|j∈[1,M]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation time, Y0(t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, and Y r,j (t) is the microseismic wave data collected by the jth underground microseismic wave detector at the tth microseismic wave propagation time, F is the mapping function of underground microseismic wave data, source position and source energy, and M is the total number of underground microseismic wave detectors; The mapping relationship between underground microseismic wave data, source position and source energy is replaced by the surface microseismic wave data, and the mapping relationship between the surface microseismic wave data, source position and source energy is obtained. The mapping relationship between the surface microseismic wave data, source position and source energy is {E(t), Y0(t)}=H({X r,i (t)|i∈[1,N]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation time, Y0(t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, and X r,i (t) is the microseismic wave data collected by the i-th surface microseismic wave detector at the t-th microseismic wave propagation time, H is the mapping function of the surface microseismic wave data and the source position and source energy, and N is the total number of surface microseismic wave detectors; The loss function of the mine disaster monitoring model is: Loss=λ1||{E(t),Y0(t)} BP -{E(t),Y0(t)} F ||1+λ2||{E(t),Y0(t)} BP -{E(t), Y0(t)} real ||1; Where Loss is the loss function, λ1 and λ2 are ||{E(t), Y0(t)} BP -{E(t), Y0(t)} F ||1 and ||{E(t), Y0(t)} BP -{E(t), Y0(t)} real ||1 weight, {E(t), Y0(t)} BP E(t), Y0(t), {E(t), Y0(t)} output by the mine disaster monitoring model F is E(t), Y0(t), {E(t), Y0(t)} calculated by underground microseismic waves real is the true value of E(t), Y0(t), ||{E(t), Y0(t)} BP -{E(t), Y0(t)} F ||1 and ||{E(t), Y0(t)} BP -{E(t), Y0(t)} real ||1 are all L1 norm forms.

2. The method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology according to claim 1 is characterized in that: Methods for calculating the earthquake source location using underground microseismic wave data include: The dynamic equation of microseismic wave propagation is established using a spherical coordinate system. The dynamic equation of microseismic wave propagation is: in, In the formula, r is the distance between the earthquake source and the underground microseismic wave detector, t is the propagation time of the microseismic wave, and Y r (t) is the microseismic wave data collected by the underground microseismic wave detector at the tth microseismic wave propagation time, Y0(tr / v) is the microseismic wave data of the earthquake source at the tr / vth microseismic wave propagation time, and v is the propagation speed of the microseismic wave; The dynamic equation of microseismic wave propagation is transformed using the earthquake source tomography positioning function to obtain the earthquake source function, which is: Y0(t)=k(r)*A p (t), where Where Y0(t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, r j is the distance from the earthquake source to the jth underground microseismic detector, A p (t) is the maximum energy value in the earthquake source tomography positioning function, k(r) is the energy diffusion compensation factor, and M is the total number of underground microseismic geophones.

3. The method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology according to claim 2 is characterized in that: Methods for calculating earthquake source energy from underground microseismic wave data include: The kinematic equation of energy flux density is constructed, and the kinematic equation of energy flux density is: In the formula, I r (t) is the energy flux density collected by the underground microseismic wave detector at the tth microseismic wave propagation time, ρ is the density of the microseismic wave propagation medium, v is the propagation speed of the microseismic wave, u is the shear modulus, r is the distance from the earthquake source to the underground microseismic wave detector, and t is the propagation time of the microseismic wave; Using T = (tr / v) Convert and get Depend on The source energy is obtained by performing an integral operation, and the source energy is: Where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation moment.

4. The method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology according to claim 3 is characterized in that: The construction of the mine disaster monitoring model includes: {X r,i (t)|i∈[1,N]} is used as the input of BP neural network, E(t) and Y0(t) are used as the output of BP neural network, and the input and output are trained by BP neural network to encapsulate the mapping relationship between surface microseismic wave data and source location and source energy, and obtain the mine disaster monitoring model; The mine disaster monitoring model is: {E(t), Y0(t)}=BP({X r,i (t)|i∈[1,N]}), where E(t) is the radiation energy data of the earthquake source at the tth microseismic wave propagation time, Y0(t) is the microseismic wave data of the earthquake source at the tth microseismic wave propagation time, and X r,i (t) is the microseismic wave data collected by the i-th surface microseismic wave detector at the t-th microseismic wave propagation time, BP is the BP neural network, and N is the total number of surface microseismic wave detectors.

5. The method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology according to claim 1 is characterized in that: The weights λ1 and λ2 are: λ1=-(1-||{E(t),Y0(t)} BP -{E(t),Y0(t)} F ||1)log(||{E(t),Y0(t)} BP -{E(t),Y0(t)} F ||1); λ2=-(1-||{E(t),Y0(t)} BP -{E(t),Y0(t)} real ||1)log(||{E(t),Y0(t)} BP -{E(t), Y0(t)} real ||1); Where λ1 and λ2 are ||{E(t), Y0(t)} BP -{E(t), Y0(t)} F ||1 and ||{E(t), Y0(t)} BP -{E(t), Y0(t)} real ||1 weight, {E(t), Y0(t)} BP E(t), Y0(t), {E(t), Y0(t)} output by the mine disaster monitoring model F is E(t), Y0(t), {E(t), Y0(t)} calculated by underground microseismic waves real is the true value of E(t), Y0(t), ||{E(t), Y0(t)} BP -{E(t), Y0(t)} F ||1 and ||{E(t), Y0(t)} BP -{E(t), Y0(t)} real ||1 are all L1 norm forms.

6. The method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology according to claim 1 is characterized in that: The surface microseismic wave data and the underground microseismic wave data are normalized.

7. The method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology according to claim 5 is characterized by: Compare Y0(t) output by the mine disaster monitoring model with the preset radiation energy threshold for disaster warning. When the Y0(t) output by the mine disaster monitoring model exceeds the preset radiation energy threshold for disaster warning, the corresponding E(t) output by the mine disaster monitoring model will be marked as the disaster source and an early warning will be issued; When Y0(t) output by the mine disaster monitoring model does not exceed the preset radiation energy threshold for disaster warning, the corresponding E(t) output by the mine disaster monitoring model will be marked as a non-disaster source.

8. A comprehensive detection and early warning system for mine disasters based on microseismic monitoring technology, characterized in that: A method for comprehensive monitoring and early warning of mine disasters based on microseismic monitoring technology as described in any one of claims 1 to 7, the system comprising: A data acquisition unit, used to collect surface microseismic wave data and underground microseismic wave data in the mining area; A data preprocessing unit, used to calculate the earthquake source position and earthquake source energy through underground microseismic wave data; and for migrating the mapping relationship between underground microseismic wave data and the location and energy of the earthquake source into the mapping relationship between surface microseismic wave data and the location and energy of the earthquake source; A deep learning unit is used to encapsulate the mapping relationship between surface microseismic wave data and the source location and source energy into a neural network to obtain a mine disaster monitoring model for determining the source location and source energy based on the surface microseismic wave data; The monitoring and early warning unit is used to monitor and warn of mine disasters in the mining area using a mine disaster monitoring model.

Images