阅读说明：本技术 一种基于bp神经网络和惯性导航的工作面调直方法 (Working face alignment method based on BP neural network and inertial navigation ) 是由 应永华 于 2021-07-21 设计创作，主要内容包括：本发明公开了一种基于BP神经网络和惯性导航的工作面调直方法,特点是将当前采煤机所处的液压支架对应的推移行程输入到训练后的BP神经网络中,得到下一个液压支架所需的推移行程,并对下一个液压支架的支架控制器发送推移指令；支架控制器根据接收到的推移指令调整液压支架到指定位置,直至完成对工作面的实时调直；优点是一方面利用BP神经网络方法,建立液压支架推移行程与惯性导航系统定位的关系,并与理想直线比较,实现采煤机定位误差处理和矫正,克服由于惯导误差导致的工作面无法真正调直现象,另一方面采煤机在割当前刀时,就通过调整液压支架对工作面进行实时调直,不但节约了人工成本,而且提高了采煤的效率和安全性。(The invention discloses a working face straightening method based on a BP (Back propagation) neural network and inertial navigation, which is characterized in that a pushing stroke corresponding to a hydraulic support where a current coal mining machine is located is input into the trained BP neural network to obtain a pushing stroke required by a next hydraulic support, and a pushing instruction is sent to a support controller of the next hydraulic support; the support controller adjusts the hydraulic support to a specified position according to the received pushing instruction until the real-time alignment of the working surface is completed; the method has the advantages that on one hand, the relation between the pushing stroke of the hydraulic support and the positioning of the inertial navigation system is established by utilizing a BP neural network method, and compared with an ideal straight line, the positioning error of the coal mining machine is processed and corrected, the phenomenon that the working surface cannot be really straightened due to inertial navigation errors is overcome, on the other hand, when the coal mining machine cuts the current cutter, the working surface is straightened in real time by adjusting the hydraulic support, the labor cost is saved, and the coal mining efficiency and the safety are improved.)

1. A working face alignment method based on a BP neural network and inertial navigation is characterized in that an infrared receiving device is installed on a support controller of each hydraulic support, an infrared transmitting device corresponding to the infrared receiving device is arranged on a coal mining machine, an inertial navigation system is installed at the center of the coal mining machine, the support controller and the inertial navigation system are communicated with an upper computer through a data transmission module, and the upper computer is used for receiving and processing data transmitted by the support controller and the inertial navigation system and sending a push command to the support controller;

the specific working face straightening method comprises the following steps:

establishing an inertial navigation coordinate system:

setting a coordinate system by taking the installation position of the inertial navigation system as an original point 0, pointing the walking direction of the coal mining machine by the original point 0 as the positive direction of an X axis, pointing the propelling direction of a working surface by the original point 0 as the positive direction of a Y axis and vertically upwards by the original point 0 as the positive direction of a Z axis；

Before the coal mining machine starts to work, an inertial navigation system collects initial coordinates of the coal mining machine and transmits the initial coordinates to an upper computer through a data transmission module, after the coal mining machine starts to work, signals received by an infrared receiving device are transmitted to the upper computer through a support controller through the data transmission module, the inertial navigation system collects positioning coordinates when the coal mining machine passes through a hydraulic support and transmits the positioning coordinates to the upper computer through the data transmission module, and the support controller transmits the pushing stroke of each hydraulic support to the upper computer through the data transmission module;

thirdly, the upper computer obtains a hydraulic support where the coal mining machine is located according to the received signals transmitted by the support controller, carries out smoothing processing on the received positioning coordinates transmitted by the inertial navigation system to obtain a positioning curve, obtains a pushing coordinate corresponding to each hydraulic support according to an inertial navigation coordinate system and the received pushing stroke transmitted by the support controller, carries out smoothing processing on all the pushing coordinates to obtain a pushing curve, translates the initial coordinate of the coal mining machine along the Y axis for the maximum pushing stroke to obtain a pushing initial coordinate, and connects the pushing initial coordinate and the pushing coordinate corresponding to the hydraulic support with the maximum pushing stroke to obtain an ideal straight line;

fourthly, constructing a BP neural network in the upper computer, inputting the positioning curve, the pushing curve and the ideal straight line obtained in the third step into the BP neural network for training to obtain the trained BP neural network;

inputting a pushing stroke corresponding to the hydraulic support where the current coal mining machine is located into the trained BP neural network to obtain a pushing stroke required by the next hydraulic support of the hydraulic support where the current coal mining machine is located, and sending a pushing instruction to a support controller of the next hydraulic support according to the pushing stroke required by the next hydraulic support;

sixthly, the support controller of the next hydraulic support adjusts the hydraulic support to a specified position according to the received push instruction, judges whether the hydraulic support is the last hydraulic support in the traveling direction of the coal mining machine according to a signal transmitted by the support controller and received by the upper computer, and returns to the fifth execution step if the hydraulic support is not the last hydraulic support in the traveling direction of the coal mining machine; if yes, executing step (c);

seventhly, judging whether the coal mining machine stops working or not, and if so, finishing the real-time alignment of the working surface; if not, the traveling direction of the coal mining machine is changed and the step (v) is returned to.

2. The working face straightening method based on the BP neural network and the inertial navigation as claimed in claim 1, wherein the BP neural network is constructed and the specific training process is as follows:

i defines a BP neural network comprising an input layer, a hidden layer and an output layer, the BP neural network comprising two stages: an information forward propagation stage and an information backward propagation stage; in the information forward transmission stage, a pushing curve enters an input layer and is transmitted to a hidden layer forward, in the hidden layer, a pushing coordinate of the pushing curve is fitted with a positioning coordinate of a positioning curve to obtain an actual coordinate, and when the actual coordinate of the output layer is inconsistent with the expected output, the information backward transmission stage is carried out; in the information back propagation stage, fitting the coordinates of the ideal straight line with the actual coordinates in the hidden layer until the output layer conforms to the expected output, wherein the expected output is that the push coordinates, the positioning coordinates and the coordinates of the ideal straight line are consistent;

and ii, inputting the pushing curve, the positioning curve and the ideal straight line into the BP neural network for training to obtain the trained BP neural network.

3. The working face straightening method based on the BP neural network and the inertial navigation according to claim 2, characterized in that the initial value of the weight of the BP neural network is randomly set between 0 and 1, and the number of nodes of each of the input layer, the hidden layer and the output layer is equal to the number of hydraulic supports.

Technical Field

The invention relates to a working face straightening method, in particular to a working face straightening method based on a BP neural network and inertial navigation.

Background

Coal always occupies the main position of energy in China, so that the safety of the coal energy is a big matter related to the national civilization. At present, the coal mining generally adopts a mechanical coal cutting technology, namely, on a working face, a coal mining machine realizes the coal cutting technology under the support of a hydraulic support, but due to the reasons of inclined working face, complex geology, non-straight arrangement of the hydraulic support, non-in-place pushing and the like, the phenomenon that the working face is not straight occurs in the coal mining process, and the major safety accidents of personnel casualties and equipment damage caused by frame falling, roof fall and the like are easily caused.

Currently, the working face alignment work of domestic coal mines is generally finished on site through manual work, so that the number and the workload of coal mining workers are increased, the coal mining efficiency is influenced, and more potential safety hazards exist; the existing method for straightening the working face through inertial navigation positioning generally comprises the steps that when a coal cutter finishes cutting a current cutter, the deviation caused by the previous cutter is corrected before cutting the next cutter, however, the inertial navigation positioning is required to have higher precision, and when the inertial navigation positioning coal cutter has larger deviation of the movement direction angle, the working face is difficult to straighten back due to limited push distance and large deviation of a hydraulic support, so that the result that the next cutter is difficult to correct is caused.

Disclosure of Invention

The technical problem to be solved by the invention is to provide a working face alignment method based on a BP neural network and inertial navigation, so that not only is the labor cost saved, but also the coal mining efficiency and safety are improved.

The technical scheme adopted by the invention for solving the technical problems is as follows: a working face alignment method based on a BP neural network and inertial navigation is characterized in that an infrared receiving device is installed on a support controller of each hydraulic support, an infrared transmitting device corresponding to the infrared receiving device is arranged on a coal mining machine, an inertial navigation system is installed at the center of the coal mining machine, the support controller and the inertial navigation system are communicated with an upper computer through a data transmission module, and the upper computer is used for receiving and processing data transmitted by the support controller and the inertial navigation system and sending a push command to the support controller;

the specific working face straightening method comprises the following steps:

establishing an inertial navigation coordinate system:

setting a coordinate system by taking the installation position of the inertial navigation system as an original point 0, pointing the walking direction of the coal mining machine by the original point 0 as the positive direction of an X axis, pointing the propelling direction of a working surface by the original point 0 as the positive direction of a Y axis and vertically upwards by the original point 0 as the positive direction of a Z axis；

Before the coal mining machine starts to work, an inertial navigation system collects initial coordinates of the coal mining machine and transmits the initial coordinates to an upper computer through a data transmission module, after the coal mining machine starts to work, signals received by an infrared receiving device are transmitted to the upper computer through a support controller through the data transmission module, the inertial navigation system collects positioning coordinates when the coal mining machine passes through a hydraulic support and transmits the positioning coordinates to the upper computer through the data transmission module, and the support controller transmits the pushing stroke of each hydraulic support to the upper computer through the data transmission module;

thirdly, the upper computer obtains a hydraulic support where the coal mining machine is located according to the received signals transmitted by the support controller, carries out smoothing processing on the received positioning coordinates transmitted by the inertial navigation system to obtain a positioning curve, obtains a pushing coordinate corresponding to each hydraulic support according to an inertial navigation coordinate system and the received pushing stroke transmitted by the support controller, carries out smoothing processing on all the pushing coordinates to obtain a pushing curve, translates the initial coordinate of the coal mining machine along the Y axis for the maximum pushing stroke to obtain a pushing initial coordinate, and connects the pushing initial coordinate and the pushing coordinate corresponding to the hydraulic support with the maximum pushing stroke to obtain an ideal straight line;

fourthly, constructing a BP neural network in the upper computer, inputting the positioning curve, the pushing curve and the ideal straight line obtained in the third step into the BP neural network for training to obtain the trained BP neural network;

inputting a pushing stroke corresponding to the hydraulic support where the current coal mining machine is located into the trained BP neural network to obtain a pushing stroke required by the next hydraulic support of the hydraulic support where the current coal mining machine is located, and sending a pushing instruction to a support controller of the next hydraulic support according to the pushing stroke required by the next hydraulic support;

sixthly, the support controller of the next hydraulic support adjusts the hydraulic support to a specified position according to the received push instruction, judges whether the hydraulic support is the last hydraulic support in the traveling direction of the coal mining machine according to a signal transmitted by the support controller and received by the upper computer, and returns to the fifth execution step if the hydraulic support is not the last hydraulic support in the traveling direction of the coal mining machine; if yes, executing step (c);

seventhly, judging whether the coal mining machine stops working or not, and if so, finishing the real-time alignment of the working surface; if not, the traveling direction of the coal mining machine is changed and the step (v) is returned to.

The BP neural network is constructed in the step IV and the specific training process is as follows:

i defines a BP neural network comprising an input layer, a hidden layer and an output layer, the BP neural network comprising two stages: an information forward propagation stage and an information backward propagation stage; in the information forward transmission stage, a pushing curve enters an input layer and is transmitted to a hidden layer forward, in the hidden layer, a pushing coordinate of the pushing curve is fitted with a positioning coordinate of a positioning curve to obtain an actual coordinate, and when the actual coordinate of the output layer is inconsistent with the expected output, the information backward transmission stage is carried out; in the information back propagation stage, fitting the coordinates of the ideal straight line with the actual coordinates in the hidden layer until the output layer conforms to the expected output, wherein the expected output is that the push coordinates, the positioning coordinates and the coordinates of the ideal straight line are consistent;

and ii, inputting the pushing curve, the positioning curve and the ideal straight line into the BP neural network for training to obtain the trained BP neural network.

And randomly setting the initial value of the weight of the BP neural network between 0 and 1, wherein the number of nodes of each layer of the input layer, the hidden layer and the output layer is equal to the number of hydraulic supports.

Compared with the prior art, the method has the advantages that on one hand, the relation between the pushing stroke of the hydraulic support and the positioning of the inertial navigation system is established by utilizing a BP neural network method, and compared with an ideal straight line, the positioning error of the coal mining machine is processed and corrected, the phenomenon that the working face cannot be straightened really due to the error of the inertial navigation system is overcome, on the other hand, when the coal mining machine cuts the front cutter, the working face is straightened in real time by adjusting the hydraulic support, so that the labor cost is saved, and the coal mining efficiency and safety are improved.

Drawings

FIG. 1 is a schematic general flow diagram of the present invention.

Detailed Description

The invention is described in further detail below with reference to the accompanying examples.

As shown in fig. 1, in a working face straightening method based on a BP neural network and inertial navigation, an infrared receiving device is installed on a support controller of each hydraulic support, an infrared transmitting device corresponding to the infrared receiving device is arranged on a coal mining machine, an inertial navigation system is installed at the center position of the coal mining machine, the support controller and the inertial navigation system are communicated with an upper computer through a data transmission module, and the upper computer is used for receiving and processing data transmitted by the support controller and the inertial navigation system and sending a push command to the support controller;

the specific working face straightening method comprises the following steps:

establishing an inertial navigation coordinate system:

setting a coordinate system by taking the installation position of the inertial navigation system as an original point 0, pointing the walking direction of the coal mining machine by the original point 0 as the positive direction of an X axis, pointing the propelling direction of a working surface by the original point 0 as the positive direction of a Y axis and vertically upwards by the original point 0 as the positive direction of a Z axis；

Before the coal mining machine starts to work, an inertial navigation system collects initial coordinates of the coal mining machine and transmits the initial coordinates to an upper computer through a data transmission module, after the coal mining machine starts to work, signals received by an infrared receiving device are transmitted to the upper computer through a support controller through the data transmission module, the inertial navigation system collects positioning coordinates when the coal mining machine passes through a hydraulic support and transmits the positioning coordinates to the upper computer through the data transmission module, and the support controller transmits the pushing stroke of each hydraulic support to the upper computer through the data transmission module;

thirdly, the upper computer obtains a hydraulic support where the coal mining machine is located according to the received signals transmitted by the support controller, carries out smoothing processing on the received positioning coordinates transmitted by the inertial navigation system to obtain a positioning curve, obtains a pushing coordinate corresponding to each hydraulic support according to an inertial navigation coordinate system and the received pushing stroke transmitted by the support controller, carries out smoothing processing on all the pushing coordinates to obtain a pushing curve, translates the initial coordinate of the coal mining machine along the Y axis for the maximum pushing stroke to obtain a pushing initial coordinate, and connects the pushing initial coordinate and the pushing coordinate corresponding to the hydraulic support with the maximum pushing stroke to obtain an ideal straight line;

fourthly, constructing a BP neural network in the upper computer, inputting the positioning curve, the pushing curve and the ideal straight line obtained in the third step into the BP neural network for training to obtain the trained BP neural network;

the BP neural network is constructed in the step IV and the specific training process is as follows:

i defines a BP neural network comprising an input layer, a hidden layer and an output layer, the BP neural network comprising two stages: an information forward propagation stage and an information backward propagation stage; in the information forward transmission stage, a pushing curve enters an input layer and is transmitted to a hidden layer forward, in the hidden layer, a pushing coordinate of the pushing curve is fitted with a positioning coordinate of a positioning curve to obtain an actual coordinate, and when the actual coordinate of the output layer is inconsistent with the expected output, the information backward transmission stage is carried out; in the information back propagation stage, fitting the coordinates of the ideal straight line with the actual coordinates in the hidden layer until the output layer is in accordance with the expected output, wherein the expected output is that the push coordinates, the positioning coordinates and the coordinates of the ideal straight line are consistent;

ii, inputting the pushing curve, the positioning curve and the ideal straight line into a BP neural network for training to obtain a trained BP neural network;

inputting a pushing stroke corresponding to the hydraulic support where the current coal mining machine is located into the trained BP neural network to obtain a pushing stroke required by the next hydraulic support of the hydraulic support where the current coal mining machine is located, and sending a pushing instruction to a support controller of the next hydraulic support according to the pushing stroke required by the next hydraulic support;

sixthly, the support controller of the next hydraulic support adjusts the hydraulic support to a specified position according to the received push instruction, judges whether the hydraulic support is the last hydraulic support in the traveling direction of the coal mining machine according to a signal transmitted by the support controller and received by the upper computer, and returns to the fifth execution step if the hydraulic support is not the last hydraulic support in the traveling direction of the coal mining machine; if yes, executing step (c);

seventhly, judging whether the coal mining machine stops working or not, and if so, finishing the real-time alignment of the working surface; if not, the traveling direction of the coal mining machine is changed and the step (v) is returned to.

In this embodiment, the initial value of the weight of the BP neural network is randomly set between 0 and 1, and the number of nodes in each of the input layer, the hidden layer, and the output layer is equal to the number of hydraulic brackets.

In this embodiment, the shearer sequentially travels on the hydraulic support according to the traveling direction.