Parameter reduction pull-in control method

Abstract

The application discloses a few-parameter pull-frame control method, and establishes a multivariable optimization relation of a pull-frame tension F, a comprehensive resistance f of a hydraulic support and a pull-frame tension duration t1 according to a dynamics principle; and an optimal solution of the multivariable optimization relation is obtained by using a genetic algorithm. The application has the advantages that for the purpose of engineering application, the pull-frame process of the hydraulic support is simplified into two processes of pull-frame and free movement, the time required for the pull-frame tension is estimated by using the genetic algorithm before the pull-frame, so that the disadvantages of segmented pull-frame and real-time adjustment are avoided, the pull-frame control precision of the hydraulic support group is improved, the straightness of the hydraulic support group is ensured, the coal wall spalling, the hydraulic support being pressed to death, the roof falling and other accidents caused by the insufficient straightness of the hydraulic support group are fundamentally avoided, the low coal output efficiency problem is solved, and the production safety of the coal mine working face and the coal mine production benefit are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of coal mining, in particular to a few-parameter support pulling control method. BACKGROUND

[0002] There are hundreds of hydraulic supports horizontally arranged in the underground working face of a coal mine. With the forward advance of the coal mining face, the hydraulic support group also needs to complete actions such as pushing and pulling. After the hydraulic support group is pushed, the telescopic rods between the hydraulic supports and the scraper conveyor are in a stretched state, and then the pulling operation is performed, i.e. the scraper conveyor is fixed, the telescopic rods are retracted, the hydraulic supports are moved closer to the scraper conveyor, and are stopped at a specified position. If all the hydraulic supports can reach the specified position accurately, the straightening of the hydraulic support group is achieved.

[0003] In this process, the straightness of the hydraulic support group directly relates to the recovery speed of the coal mining face and the safety support of the surrounding rock. If the straightness of the hydraulic support group is not enough, it may cause uneven force of the hydraulic support group, thereby causing the hydraulic support to tilt, etc. Especially in soft coal seams, broken coal seams and other geological conditions, it is easy to cause coal wall spalling, hydraulic support crushing, roof falling and other accidents. At the same time, if the straightness of the hydraulic support group is not enough, the scraper conveyor will be severely bent, thereby causing the cutting depth of the coal mining machine to be uneven when recovering the coal wall, resulting in low coal production efficiency and seriously affecting the coal production benefit.

[0004] The key to affecting the straightening of the hydraulic support group is the pulling process, i.e. the pulling control method of the hydraulic support group. However, due to the short pulling distance of the hydraulic support group and the small actual space, it is impossible to install data acquisition equipment such as pressure sensors, so in the pulling process of the hydraulic support group, it is impossible to obtain necessary data such as the pulling force of the hydraulic support group, the friction force between the support base and the floor, and the friction force between the top beam and the roof, and it is difficult to establish an accurate control model. Moreover, the pulling time is continuous and the distance is short, so it is also impossible to achieve accurate pulling control through online inference algorithms. This results in that even today when computer control is highly developed, manual operation is still needed to achieve the pulling action of the hydraulic support group.

[0005] At present, the more advanced hydraulic support group pulling control methods include setting experience value method, online two-point extrapolation or three-point extrapolation method, and Kalman filtering method. However, since the state of the floor is different every time the coal mining face advances one cut, the setting experience value method is often invalid. In addition, the online two-point extrapolation or three-point extrapolation method divides the hydraulic support group pulling process into two or three sections, the front part of the pulling process is used to estimate the pulling force F and the comprehensive resistance f, and the rear part is used to establish a model according to the estimated pulling force F and the comprehensive resistance f to pull the hydraulic support group into place. However, since the space distance of the pulling is very short and the speed is very fast, it is difficult to meet the requirements of the segmented pulling, so the method is usually difficult to implement in practice. The Kalman filtering method is used to estimate the pulling force in real time and dynamically adjust the pulling pressure. However, since the single-speed valve or double-speed valve is commonly used in the hydraulic support, the pressure is adjusted according to the opening ratio; at the same time, the hydraulic system cannot withstand the impact of pressure pulse, so the pulse type liquid supply cannot be used to adjust the pressure. Therefore, the Kalman filtering method cannot be implemented in practice. SUMMARY

[0006] The present application aims to provide a few-parameter pulling control method.

[0007] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0008] The few-parameter pulling control method provided by the present application comprises the following steps:

[0009] S1, a multivariate optimization relationship of the pulling force F, the hydraulic support comprehensive resistance f and the pulling force duration t1 is established according to the dynamics principle;

[0010] S2, the optimal solution of the multivariate optimization relationship is obtained by using the genetic algorithm.

[0011] Further, the multivariate optimization relationship is as follows:

[0012]

[0013] wherein, is the total distance that needs to be pulled by the hydraulic support; m is the mass of the hydraulic support;

[0014] Further, when solving the multivariate optimization relationship, the F, f need to meet the following solution domain conditions:

[0015]

[0016] wherein, is the maximum value of the pulling force; is the maximum value of the hydraulic support comprehensive resistance obtained according to experience; The maximum action time of the pull frame pulling force obtained according to experience.

[0017] Further, the solution process of the genetic algorithm is:

[0018] S2.1, using 16bit coding mode, the solution domain condition of F, f is binary coded;

[0019] S2.2, randomly generate N groups of binary values of F, f;

[0020] S2.3, according to the adaptive evaluation function, select the N / 10 binary values with the minimum evaluation value as the next generation genetic individuals;

[0021] S2.4, cross rate and mutation rate respectively on the next generation of genetic individuals to cross and mutate, and obtain N / 10 next generation genetic individuals again;

[0022] S2.5, repeat S2.3 and S2.4 until a unique genetic individual is screened out as an optimized solution of the multivariate optimization relationship;

[0023] S2.6, execute S2.2 to S2.5 for multiple times to obtain multiple optimized solutions of the multivariate optimization relationship;

[0024] S2.7, the optimized solution of F closest to in the multiple optimized solutions is the optimal solution of the multivariate optimization relationship, so as to determine ;

[0025] S2.8, according to the value of control the duration of the pull frame pulling force.

[0026] Further, the genetic algorithm is used to solve the multivariate optimization relationship. ; the genetic algorithm is used to solve the multivariate optimization relationship. .

[0027] Further, the adaptive evaluation function is

[0028] .

[0029] ​​The present application has the advantages that for the purpose of engineering application, the pulling process of hydraulic support is simplified into two processes of pulling and free movement, before pulling, the time required for pulling force is estimated by using genetic algorithm, thereby avoiding the disadvantages of segmented pulling and real-time adjustment, improving the pulling control precision of hydraulic support group, thereby guaranteeing the straightness of hydraulic support group, fundamentally avoiding the problems of coal wall spalling, hydraulic support being pressed to death, roof falling and low coal production efficiency caused by insufficient straightness of hydraulic support group, and ensuring the production safety of coal mine working face and the production benefit of coal mine. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is the hydraulic support pulling control space structure diagram of the present application.

[0031] Figure 2 is the hydraulic support pulling process diagram of the present application.

[0032] Figure 3 is the hydraulic support pulling process modeling diagram of the present application.

[0033] Figure 4 is the method flow chart of the present application.

[0034] Figure 5 is the genetic algorithm solving process flow chart of the method of the present application. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application will be clearly and completely described below, obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0036] As shown in Figure 1 , wherein Figure 1 -a is a side view, and 1-b is a top view. After the hydraulic support 1 completes pushing, the telescopic rod 2 is in the stretched state; then the pulling operation is performed, that is, the scraper conveyor 3 is fixed, the telescopic rod 2 is retracted, and the hydraulic support 1 is close to the scraper conveyor 3. When the pulling is completed, the base of the hydraulic support 1 is not completely close to the scraper conveyor 3, but stops at the specified stopping point 4.

[0037] As shown in Figure 2 , it is a hydraulic support 1 pulling process diagram, wherein m is the mass of the hydraulic support, F is the pulling force, f is the comprehensive resistance of the hydraulic support, S1 is the running displacement of the hydraulic support under the action of the pulling force F, S2 is the free sliding distance of the hydraulic support due to inertia, and the sum of S1 and S2 is the total distance S m that the hydraulic support needs to pull.

[0038] As Figure 3 shown, the pull frame process is modeled, that is, in the action time t1 of the pull frame force F, the hydraulic support running speed reaches v1, and then after the time t2, the hydraulic support running speed will be 0, and stops at the specified stop point 4. In order to realize accurate pull frame, it is necessary to determine the action time t1 of the pull frame force F, so as to ensure the straightness of the hydraulic support group. For this purpose, the few-parameter pull frame control method of the application comprises the following steps:

[0039] S1, a multivariate optimization relationship of the pull frame force F, the comprehensive resistance f of the hydraulic support and the pull frame force duration t1 is established according to the dynamics principle;

[0040] Since the condition of the bottom plate after the hydraulic support is pushed is relatively single, and the space distance of the hydraulic support pull frame is short, the comprehensive resistance f of the bottom plate to the hydraulic support can be considered as a constant value. And because the actual space is small, it is impossible to install a mechanical sensor to obtain the real-time value of the pull frame force F, and at the same time, the pull frame process is very short in time, generally for a few seconds, so the pull force F can be considered as a constant.

[0041] According to the dynamics principle, we have:

[0042] (1)

[0043] (2)

[0044] (3)

[0045] (4)

[0046] According to formula (1) and formula (2), we have:

[0047] (5)

[0048] Then we have

[0049] (6)

[0050] In formula (6), m is a known quantity, and the sum of S1 and S2 is the total distance S that the hydraulic support needs to pull m , which is also a known quantity. ,F,f are unknown quantities, so the problem of determining the action time t1 of the pull frame force F can be converted into a multivariate variable optimization problem of F, f. According to formula (6), the multivariate optimization relationship is established as:

[0051] (7)

[0052] wherein, ,F,f need to meet the following solution domain conditions:

[0053] (8)

[0054] wherein, is the maximum value of the pull force, and is a known quantity; is the maximum value of the comprehensive resistance of the hydraulic support obtained empirically; is the maximum action time of the pull force obtained empirically.

[0055] S2, the optimal solution of the multivariable optimization relationship is obtained by using a genetic algorithm.

[0056] The solution process of the genetic algorithm is as follows:

[0057] S2.1, the solution domain conditions of,F,f are binary coded; that is, the decimal numbers of,F,f are converted into binary numbers, and the conversion formula is:

[0058] (9)

[0059] wherein, may be , and ; is a binary number; is a decimal number.

[0060] S2.2, N groups of binary values of,F,f are randomly generated;

[0061] S2.3, the N / 10 binary values with the minimum evaluation value are selected as the next generation of genetic individuals according to the fitness evaluation function; the fitness evaluation function is:

[0062] (10)

[0063] S2.4, the next generation of genetic individuals are crossed and mutated at a crossover rate and a mutation rate respectively, and N / 10 next generation of genetic individuals are obtained again; wherein ; .

[0064] S2.5, the steps S2.3 and S2.4 are repeatedly executed until a unique genetic individual is screened out, which is an optimal solution of the multivariable optimization relationship;

[0065] S2.6, repeatedly performing S2.2 to S2.5 to obtain multiple optimization solutions of the multivariate optimization relationship; in general, at least 10 times of repetition is required to obtain 10 optimization solutions of the multivariate optimization relationship.

[0066] S2.7, the optimization solution with the F value closest to in the 10 optimization solutions is the optimal solution of the multivariate optimization relationship, so that ;

[0067] S2.8, according to the value, the duration of the pull force can be accurately controlled, so as to ensure the straightness of the hydraulic support group.

Claims

1. A few-parameter pull architecture control method, characterized in that: The method comprises the following steps: S1, a multivariate optimization relationship of a pull frame pulling force F, a comprehensive resistance f of a hydraulic support and a duration t1 of the pulling force is established according to a kinetic principle; The multivariate optimization relationship is: wherein, is the total distance needed to pull the hydraulic support; m is the mass of the hydraulic support; S2, an optimal solution of the multivariate optimization relationship is obtained by using a genetic algorithm.

2. The reduced parameter pull architecture control method of claim 1, wherein: In solving the multivariable optimization relationship, the F,f need to meet the following solution domain conditions: wherein, is the maximum value of the support force; is the maximum value of the comprehensive resistance of the hydraulic support obtained empirically; is the maximum action time of the support force obtained empirically.

3. The reduced parameter pull architecture control method of claim 1, wherein: The solving process of the genetic algorithm is: S2.1, using 16-bit encoding, the domain of the solution of the equation the binary encoding of the domain of solution of the equation S2.2, randomly generate N sets of binary values; S2.3, N / 10 binary values with the minimum evaluation value are selected as the next generation genetic individuals according to the adaptive evaluation function; S2.4, at a crossover rate and a mutation rate crossing and mutating the next generation genetic individuals respectively, to obtain N / 10 next generation genetic individuals again; S2.5, the steps S2.3 and S2.4 are repeatedly executed until a unique genetic individual is screened out as an optimization solution of the multivariate optimization relationship; S2.6, the steps S2.2 to S2.5 are repeatedly executed for multiple times to obtain multiple optimization solutions of the multivariate optimization relationship; S2.7, F closest to among multiple optimization solutions the optimization solution is the optimal solution of the multivariate optimization relationship, thereby determining ; S2.8, according to The value controls the duration of the pull frame pull force.

4. The reduced parameter pull architecture control method of claim 3, wherein: The ; the .

5. The reduced parameter pull architecture control method of claim 3, wherein: The adaptive evaluation function is 。

Images