Tunnel Ventilation Environment Optimization Method, System, Storage Medium and Electronic Device

Abstract

The tunnel ventilation environment optimization method, system, storage medium and electronic device of the present invention belong to the field of tunnel construction technology. The method comprises: S1: establishing a PID control system; S2: obtaining ventilation environment parameters of the tunnel; S3: optimizing the control parameters of the PID control system in real time using an RBF neural network to obtain an optimized PID control system; S4: inputting the ventilation environment parameters into the optimized PID control system, dynamically adjusting the drive system of the tunnel ventilation equipment through the optimized PID control system, thereby optimizing the tunnel ventilation environment. The present invention overcomes the limitations of the traditional PID control system in dealing with nonlinearity, time-varying and uncertainty, improves the control accuracy and robustness of the PID control system, and thus improves the accuracy of its control of the drive system of the tunnel ventilation equipment.

Description

Technical Field

[0001] The present invention belongs to the technical field of tunnel construction, and relates to ventilation technology in the process of tunnel construction, specifically to a method, system, storage medium and electronic equipment for optimizing the ventilation environment of a tunnel. Background Art

[0002] Tunnel construction in high-altitude areas faces specific environmental challenges, including low oxygen environment, harmful gases and large amounts of dust. The low pressure state in high-altitude environments will lead to a thin oxygen supply, which will increase the risk of construction workers suffering from hypoxemia. At the same time, during the construction process, such as drilling and blasting operations, mechanical excavation, etc., a large amount of dust and harmful gases will be released. If not handled in a timely and effective manner, it will seriously affect the health of construction workers and the safety of construction.

[0003] Traditional tunnel construction ventilation, dust removal, oxygen supply, etc. mostly adopt relatively simple automatic control technology. See the patent application document with application number CN201710648851.0, which discloses a tunnel intelligent ventilation control method, which calculates the corresponding deviation and deviation change rate through the pollutant concentration, wind speed, and traffic volume parameters fed back by the on-site detector, and then adjusts the proportional coefficient Kp, integral action coefficient Ki, and differential action coefficient Kd according to the preset fuzzy control rules. The control frequency of the frequency converter is obtained through PID operation, and the jet fan is controlled to operate with variable frequency. The pollutant concentration and wind speed parameters are cyclically fed back and the current control frequency is continuously calculated and output to achieve ventilation on demand. This patent document uses a PID controller to automatically adjust the frequency of the frequency converter to achieve ventilation on demand. However, for high-altitude areas, the oxygen content, harmful gas content, and dust content are very high during the construction process, and real-time monitoring, adjustment and optimization are required. The traditional PID control system has poor anti-interference ability and oscillation, which makes it impossible to adjust the control parameters of the PID control system in real time, and thus cannot be applied to the tunnel ventilation environment adjustment in high-altitude areas. Summary of the invention

[0004] As described in the background technology, the traditional PID control system has poor anti-interference ability, oscillation phenomenon, and cannot adjust the control parameters of the PID control system in real time, resulting in the technical problem that it is not suitable for tunnel ventilation environment regulation in high-altitude areas. To solve this technical problem, the present invention proposes a tunnel ventilation environment optimization method, system, storage medium and electronic device.

[0005] The present invention utilizes the approximation function of the RBF neural network to identify the dynamic characteristics of the PID control system, and dynamically adjusts the control parameters (proportional coefficient K p , integral action coefficient K i and the differential action coefficient K d), determine the optimal values ​​of the control parameters of the PID control system, overcome the limitations of traditional PID control systems in dealing with nonlinearity, time-varying and uncertainty, improve the control accuracy and robustness of the PID control system, thereby improving the accuracy of its control of the drive system of tunnel ventilation equipment, and can be better applied to tunnel ventilation environment regulation in high-altitude areas.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The method for optimizing the tunnel ventilation environment of the present invention comprises the following steps:

[0008] S1: Establish PID control system;

[0009] S2: Obtain the ventilation environment parameters of the tunnel;

[0010] S3: Utilize RBF neural network to iteratively optimize the control parameters of the PID control system to obtain an optimized PID control system;

[0011] S4: The ventilation environment parameters are input into the optimized PID control system, and the driving system of the tunnel ventilation equipment is dynamically adjusted through the optimized PID control system, so as to optimize the tunnel ventilation environment.

[0012] It is further defined that step S1 specifically includes:

[0013] S2.1: Obtain the initial ventilation environment parameters of the tunnel;

[0014] S2.2: De-noising the initial ventilation environment parameters to form ventilation environment parameters.

[0015] It is further defined that step S2 specifically includes: according to a set time threshold, using an RBF neural network to iteratively optimize control parameters of the PID control system to obtain an optimized PID control system.

[0016] It is further defined that in step S3, the optimized PID control system is:

[0017]

[0018] In the formula, For the The ventilation increment for iterations, in units: ; For the first ( -1) The ventilation increment of the iteration, unit: ; is the change of the deviation signal, dimensionless; is the current error value, dimensionless; is the differential value of the error, dimensionless; is the proportionality coefficient, dimensionless; is the adjustment of the proportionality coefficient, dimensionless; is the integral action coefficient, dimensionless; is the adjustment of the integral action coefficient, dimensionless; is the differential action coefficient, dimensionless; It is the adjustment amount of the differential action coefficient and is dimensionless.

[0019] It is further defined that the adjustment amount of the proportional coefficient , adjustment of the integral action coefficient and the adjustment of the differential action coefficient The corresponding calculation formulas are:

[0020]

[0021] In the formula, is the learning rate of the proportional coefficient, dimensionless; for Deviation signal of the time step, dimensionless; For the tunnel Ventilation environment parameter within the time step, dimensionless; is the output value of the RBF neural network, dimensionless; is the change in the deviation signal, dimensionless; is the learning rate of the integral action coefficient, dimensionless; is the current error value, dimensionless; is the learning rate of the differential action coefficient, dimensionless; is the differential value of the error, dimensionless.

[0022] Further defined, in step S2.1, the initial ventilation environment parameters include oxygen concentration, dust concentration, NO 2 Concentration, CO 2 concentration, CO concentration, temperature and humidity, SO 2 Concentration and / or H 2 S concentration.

[0023] It is further defined that in step S2.2, the initial ventilation environment parameters are denoised by a normalization method.

[0024] The tunnel ventilation environment optimization system of the present invention is applied to the above-mentioned tunnel ventilation environment optimization method, comprising:

[0025] System establishment module: used to establish a PID control system;

[0026] Parameter acquisition module: used to obtain the ventilation environment parameters of the tunnel;

[0027] System optimization module: used to iteratively optimize the control parameters of the PID control system using the RBF neural network to obtain an optimized PID control system;

[0028] And the environmental optimization module: it is used to input the ventilation environment parameters into the optimized PID control system, and dynamically adjust the driving system of the tunnel ventilation equipment through the optimized PID control system, so as to optimize the tunnel ventilation environment.

[0029] The present invention provides a storage medium storing a program file, wherein the program file is executed to implement the above-mentioned tunnel ventilation environment optimization method.

[0030] The present invention provides an electronic device, comprising a processor and a memory coupled to each other, wherein:

[0031] The memory is used to store the above-mentioned method for optimizing the tunnel ventilation environment;

[0032] The processor is used to execute program instructions stored in the memory.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The tunnel ventilation environment optimization method of the present invention utilizes the approximation function of the RBF neural network to identify the dynamic characteristics of the PID control system, and dynamically adjusts the control parameters (proportional coefficient K p , integral action coefficient K i and the differential action coefficient K d ), determine the optimal values ​​of the control parameters of the PID control system, overcome the limitations of traditional PID control systems in dealing with nonlinearity, time-varying and uncertainty, improve the control accuracy and robustness of the PID control system, thereby improving the accuracy of its control of the drive system of tunnel ventilation equipment, and can be better applied to tunnel ventilation environment regulation in high-altitude areas.

[0035] The present invention optimizes and adjusts the control parameters of the PID control system through the RBF neural network, realizes the real-time dynamic adjustment of the ventilation volume, dust removal and oxygen supply during the tunnel ventilation process, and meets the needs of the tunnel construction environment in an optimal way. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of a method for optimizing tunnel ventilation environment according to the present invention;

[0037] Figure 2 A schematic diagram of a tunnel ventilation environment optimization system according to the present invention;

[0038] Figure 3 For different learning rates Simulation diagram of ventilation volume controlled by optimized PID control system under the following values;

[0039] Figure 4 For different momentum factors Simulation diagram of ventilation volume controlled by optimized PID control system under the following values;

[0040] Figure 5 For different , and Simulation diagram of ventilation volume controlled by optimized PID control system under different values. DETAILED DESCRIPTION

[0041] The technical solution of the present invention is further explained below in conjunction with the accompanying drawings and embodiments, but the present invention is not limited to the implementation modes described below.

[0042] See also Figure 1 The tunnel ventilation environment optimization method of the present invention comprises the following steps:

[0043] S1: Establish PID control system.

[0044] S2: Obtain the ventilation environment parameters of the tunnel.

[0045] Specifically, step S2 includes:

[0046] S2.1: Obtain initial ventilation environment parameters of the tunnel; S2.2: De-noise the initial ventilation environment parameters to form ventilation environment parameters.

[0047] S3: Use RBF neural network to iteratively optimize the control parameters of the PID control system to obtain an optimized PID control system.

[0048] Specifically, step S3 is: according to a set time threshold, the control parameters of the PID control system are iteratively optimized using the RBF neural network to obtain an optimized PID control system, and the time threshold refers to the time interval between two adjacent iterations.

[0049] Among them, the RBF neural network is a three-layer single hidden layer forward neural network consisting of an input layer, a hidden layer, and an output layer. Unlike other neural networks, the hidden layer neurons of the RBF neural network use the basis function (RBF) as the activation function, and its output value is only related to the distance between the input values. The input layer only serves to transmit signals, and the hidden layer adjusts the input parameters through nonlinear mapping, and each neuron is a basis function.

[0050] The output layer of the RBF neural network is adjusted by changing the linear weights to complete the response to the activation signal of the input layer. The representation of the output layer includes perception units, which refer to radial basis function units. Their main function is to map the input data to a high-dimensional space. The high-dimensional space can make the data more linearly separable or easier to regress.

[0051] RBF neural network has multi-dimensional nonlinear mapping ability and generalization ability. Theoretically, it is proved that RBF neural network can approximate any nonlinear function with arbitrary precision and has the characteristic of unique optimal approximation. It has the advantages of simple parameter adjustment, fast convergence speed and strong anti-noise ability.

[0052] Tuning Index of RBF Neural Network for:

[0053]

[0054] In the formula, for Deviation signal at time step.

[0055] The input of the PID control system is:

[0056]

[0057] In the formula, is the change of the deviation signal, dimensionless; for( ) is the deviation signal of the time step, dimensionless; is the current error value, dimensionless; is the differential value of the error, dimensionless; for Deviation signal of the time step, dimensionless; for Deviation signal at time step, dimensionless.

[0058] The control parameters of the PID control system are determined by the gradient descent method, where the control parameters include the proportional coefficient K p , integral action coefficient K i and the differential action coefficient K d , the adjustment amount is:

[0059]

[0060] In the formula, is the adjustment of the proportionality coefficient, dimensionless; is the learning rate of the proportional coefficient, dimensionless; is the deviation signal of the regulation quantity, dimensionless; For the tunnel Ventilation environment parameter within the time step, dimensionless; is the output value of the RBF neural network, dimensionless; is the adjustment of the integral action coefficient, dimensionless; is the learning rate of the integral action coefficient, dimensionless; is the adjustment of the differential action coefficient, dimensionless; is the learning rate of the differential action coefficient, dimensionless; is the deviation change, dimensionless; is the current error value, dimensionless; is the differential value of the error, dimensionless.

[0061] Combining the expression of the control index and the ventilation environment parameters of the tunnel, the following formula can be obtained:

[0062]

[0063] In the formula, For the The ventilation increment for iterations, in units: ; For the The error value of the deviation signal of the iteration, dimensionless; For the first ) iteration deviation signal error value, dimensionless; For the first ) iteration deviation signal error value, dimensionless; ; is the proportionality coefficient, dimensionless; is the integral action coefficient, dimensionless; is the differential action coefficient, dimensionless.

[0064] Combined with the characteristics of tunnel construction ventilation environment, the ventilation increment is set as a multiple of the actual ventilation volume, then:

[0065]

[0066] Substituting equation (3) into equation (5), we get the optimized PID control system:

[0067]

[0068] In the formula, For the The ventilation increment for iterations, in units: ; For the first ( -1) The ventilation increment of the iteration, unit: ; is the change of the deviation signal, dimensionless; is the current error value, dimensionless; is the differential value of the error, dimensionless; is the proportionality coefficient, dimensionless; is the adjustment of the proportionality coefficient, dimensionless; is the integral action coefficient, dimensionless; is the adjustment of the integral action coefficient, dimensionless; is the differential action coefficient, dimensionless; It is the adjustment amount of the differential action coefficient and is dimensionless.

[0069] S4: Input the ventilation environment parameters into the optimized PID control system, and dynamically adjust the driving system of the tunnel ventilation equipment through the optimized PID control system, so as to optimize the tunnel ventilation environment; wherein the driving system of the tunnel ventilation equipment is the frequency converter used by the fan for ventilation, the frequency converter used by the fan for dust removal, etc.

[0070] In the present invention, the initial ventilation environment parameters include oxygen concentration, dust concentration, NO 2 Concentration, CO 2 concentration, CO concentration, temperature and humidity, SO 2 Concentration and / or H 2 S concentration.

[0071] In the present invention, the initial ventilation environment parameters are denoised by a normalization method.

[0072] See also Figure 2 The tunnel ventilation environment optimization system of the present invention is applied to the above-mentioned tunnel ventilation environment optimization method, comprising:

[0073] System establishment module: used to establish a PID control system;

[0074] Parameter acquisition module: used to obtain the ventilation environment parameters of the tunnel;

[0075] System optimization module: used to iteratively optimize the control parameters of the PID control system using the RBF neural network to obtain an optimized PID control system;

[0076] And the environmental optimization module: it is used to input the ventilation environment parameters into the optimized PID control system, and dynamically adjust the driving system of the tunnel ventilation equipment through the optimized PID control system, so as to optimize the tunnel ventilation environment.

[0077] Among them, in the present invention, the tunnel ventilation environment optimization system and the tunnel ventilation environment optimization method correspond to each other, and the system establishment module, parameter acquisition module, system optimization module and environment optimization module are specifically described in the tunnel ventilation environment optimization method part.

[0078] The present invention provides a storage medium storing a program file, wherein the program file is executed to implement the above-mentioned tunnel ventilation environment optimization method.

[0079] The storage medium in the present invention may specifically include random access memory (RAM), internal memory, read-only memory (ROM), programmable ROM, erasable programmable ROM, register, hard disk, removable disk or CD-ROM. It should be noted that those skilled in the art may specifically select the form and type of storage medium according to actual use requirements, and the present invention does not make further specific limitations.

[0080] The present invention provides an electronic device, comprising a processor and a memory coupled to each other, wherein:

[0081] The memory is used to store the above-mentioned method for optimizing the tunnel ventilation environment;

[0082] The processor is used to execute program instructions stored in the memory.

[0083] The electronic device in the present invention includes any electronic device that can execute program instructions, such as a computer, a mobile terminal or a wearable device.

[0084] The present invention uses the learning rate Affects the reaction speed of control parameter adjustment. If the value is too small, the number of iterations required to control the model parameters to achieve a better control effect will increase. If the value is too large, the local minimum value will not converge, and the local optimal value will not be obtained. The magnitude of the change in the momentum of the trend, which is related to the learning rate Together they affect the iteration speed, when the momentum factor If the value is too small, the rate at which the control model parameters reach the optimal value is lower than the rate at which the PID control system changes. The PID control system based on the RBF neural network cannot play a timely adjustment role, which will cause chaos in the entire PID control system. If the value is too large, the control model parameters are prone to overshoot, causing the PID control system to oscillate. , learning rate of integral action coefficient and the learning rate of the derivative action coefficient The degree of change of the control parameters that directly affects the PID control system is the degree of influence of the RBF neural network on the control parameters of the PID control system. When the initial values ​​of the control parameters of the PID control system are determined, , and The smaller the value of is, the closer the control effect is to the original PID control system. , and The larger the value of is, the more sensitive the control parameter adjustment of the PID control system is, and the oscillation trend is also hidden in it. The PID initial parameter of the RBF neural network PID control model is =0.3, =-6, =0.06, assuming that the number of radial basis functions is 7, and the center of the radial basis function is:

[0085]

[0086] Depend on Figure 3 , Figure 4 and Figure 5 It is known that when , momentum factor hour, , and When both are 0.1, it is more suitable for tunnel construction ventilation system. Figure 5 In for , and A general term for .

[0087] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments can still be modified, or part or all of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the present invention.

Claims

1. A method for optimizing tunnel ventilation environment, characterized in that: The following steps are involved: S1: Establish PID control system; S2: Obtain the ventilation environment parameters of the tunnel; S3: Utilize RBF neural network to iteratively optimize the control parameters of the PID control system to obtain an optimized PID control system; wherein the optimized PID control system is: In the formula, For the The ventilation increment for iterations, in units: ; For the first ( -1) The ventilation increment of the iteration, unit: ; is the change in the deviation signal, dimensionless; is the current error value, dimensionless; is the differential value of the error, dimensionless; is the proportionality coefficient, dimensionless; is the adjustment of the proportionality coefficient, dimensionless; is the integral action coefficient, dimensionless; is the adjustment of the integral action coefficient, dimensionless; is the differential action coefficient, dimensionless; is the adjustment of the differential action coefficient, dimensionless; S4: The ventilation environment parameters are input into the optimized PID control system, and the driving system of the tunnel ventilation equipment is dynamically adjusted through the optimized PID control system, so as to optimize the tunnel ventilation environment.

2. The method for optimizing tunnel ventilation environment according to claim 1, characterized in that: The step S2 specifically includes: S2.1: Obtain the initial ventilation environment parameters of the tunnel; S2.2: De-noising the initial ventilation environment parameters to form ventilation environment parameters.

3. The method for optimizing tunnel ventilation environment according to claim 1, characterized in that: The step S3 specifically includes: according to the set time threshold, using the RBF neural network to iteratively optimize the control parameters of the PID control system to obtain an optimized PID control system.

4. The method for optimizing tunnel ventilation environment according to claim 1, characterized in that: The adjustment amount of the proportional coefficient , adjustment of the integral action coefficient and the adjustment of the differential action coefficient The corresponding calculation formulas are: In the formula, is the learning rate of the proportional coefficient, dimensionless; for Deviation signal of the time step, dimensionless; For the tunnel Ventilation environment parameter within the time step, dimensionless; is the output value of the RBF neural network, dimensionless; is the change in the deviation signal, dimensionless; is the learning rate of the integral action coefficient, dimensionless; is the current error value, dimensionless; is the learning rate of the differential action coefficient, dimensionless; is the differential value of the error, dimensionless.

5. The method for optimizing tunnel ventilation environment according to claim 2, characterized in that: In step S2.1, the initial ventilation environment parameters include oxygen concentration, dust concentration, NO2 concentration, CO2 concentration, CO concentration, temperature and humidity, SO2 concentration and / or H2S concentration.

6. The method for optimizing tunnel ventilation environment according to claim 2, characterized in that: In the step S2.2, the initial ventilation environment parameters are denoised by a normalization method.

7. A tunnel ventilation environment optimization system, applied to the tunnel ventilation environment optimization method according to claim 1, characterized in that: include: System establishment module: used to establish a PID control system; Parameter acquisition module: used to obtain the ventilation environment parameters of the tunnel; System optimization module: used to iteratively optimize the control parameters of the PID control system using the RBF neural network to obtain an optimized PID control system; wherein the optimized PID control system is: In the formula, For the The ventilation increment for iterations, in units: ; For the first ( -1) The ventilation increment of the iteration, unit: ; is the change in the deviation signal, dimensionless; is the current error value, dimensionless; is the differential value of the error, dimensionless; is the proportionality coefficient, dimensionless; is the adjustment of the proportionality coefficient, dimensionless; is the integral action coefficient, dimensionless; is the adjustment of the integral action coefficient, dimensionless; is the differential action coefficient, dimensionless; is the adjustment amount of the differential action coefficient, dimensionless; And the environmental optimization module: it is used to input the ventilation environment parameters into the optimized PID control system, and dynamically adjust the driving system of the tunnel ventilation equipment through the optimized PID control system, so as to optimize the tunnel ventilation environment.

8. A storage medium, characterized in that: A program file is stored, and the program file is executed to implement the tunnel ventilation environment optimization method according to any one of claims 1 to 6.

9. An electronic device, characterized in that: The invention comprises a processor and a memory coupled to each other, wherein: The memory is used to store the method for optimizing the tunnel ventilation environment according to any one of claims 1 to 6; The processor is used to execute program instructions stored in the memory.

Images