Self-energy-supply type hoisting skip bucket stress monitoring system and method based on data inversion

Abstract

The invention discloses a self-energized lifting skip bucket stress monitoring system and method based on data inversion, and the method comprises the steps: obtaining the stress distribution conditions of a skip bucket in key regions under different working conditions through a finite element analysis method, and determining the key region needing to be monitored; acquiring transmission characteristics among different stress to-be-measured points in the key areas by using a long-short-term memory network, respectively taking stress data acquired under different working conditions as input, and predicting overall stress data of each key area of the skip bucket under different working conditions by using the long-short-term memory neural network; according to the vibration frequency of the skip bucket under different working conditions, the array type piezoelectric vibration energy collection module and the energy management module collect and manage the vibration energy of the skip bucket under different working conditions based on the piezoelectric effect; the current operation speed of the skip bucket is obtained by using a skip bucket operation speed sensing strategy based on voltage frequency, and periodic working and sleep modes of nodes at different monitoring positions are switched according to stress distribution characteristics of the skip bucket under different working conditions.

Description

Technical Field

[0001] The invention relates to a self-powered hoisting bucket stress monitoring system and method based on data inversion, belonging to the technical field of stress monitoring node deployment of hoisting buckets. Background Art

[0002] As an important load-bearing component of the mine hoisting system, the hoisting skip is prone to failure and deformation of the supporting structure due to long-term heavy-load operation in harsh working conditions, which in turn affects the safety and efficiency of coal mine hoisting. The stress of the hoisting skip can directly reflect its deformation and load, and is a key parameter for realizing the health status monitoring of the hoisting skip. Therefore, it is very necessary to study the stress monitoring method of the hoisting skip.

[0003] However, most of the existing methods for monitoring the stress of the lifting bucket use large-scale wireless sensor network nodes. This method has the following main problems: 1) Due to the large scale of the wireless sensor nodes, the staff needs to perform long-term additional maintenance; 2) The working environment of the lifting bucket is very harsh, and it is difficult to use wired power supply for the nodes. The battery needs to be replaced regularly, which is time-consuming and labor-intensive; 3) When the number of bucket stress monitoring points reaches a certain level, the monitoring area will reach an oversaturated monitoring state. At this time, the monitoring effect will no longer be significantly improved, and the monitoring efficiency will be reduced. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a self-powered lifting bucket stress monitoring system and method based on data inversion, which can realize real-time acquisition of stress data of the bucket under different working conditions while optimizing the number of bucket sensors; at the same time, considering that there is vibration energy in the bucket during operation, the present invention utilizes an array piezoelectric vibration energy collection module and an energy management module to realize the collection and management of the vibration energy of the bucket under different operating conditions, thereby replenishing energy for the stress monitoring node; secondly, based on the above-mentioned energy management module and energy collection module, the present invention discloses a bucket operation speed perception strategy based on voltage frequency, which realizes real-time perception of the current speed of the bucket, and based on the stress distribution characteristics of the bucket under different operating conditions (different loads, speeds), realizes the autonomous switching of periodic working and sleep modes of nodes at different monitoring positions.

[0005] Preferably, the present invention provides a self-powered lifting bucket stress monitoring system based on data inversion, comprising a plurality of stress monitoring nodes, an energy collection module and an energy management module, wherein the plurality of stress monitoring nodes are installed on the bucket, and the energy collection module is connected to the energy management module; wherein the energy management module comprises a bracket, a plurality of piezoelectric vibrators, a plurality of piezoelectric cantilever beams and a plurality of mass blocks of different weights, the bucket, the bracket, the piezoelectric cantilever beam and the mass block are fixedly connected in sequence, the piezoelectric cantilever beams are fixedly provided with piezoelectric vibrators, the piezoelectric vibrators are connected to the energy management module, and the energy management module is connected to the stress monitoring nodes.

[0006] Preferably, a method for self-powered hoisting bucket stress monitoring based on data inversion, using a system for self-powered hoisting bucket stress monitoring based on data inversion, performs:

[0007] Step A, using a finite element analysis method to obtain stress distribution in key areas of the bucket under different working conditions, and determine the key areas that need to be monitored, wherein the different working conditions include no-load descent, full-load acceleration-deceleration ascent, and full-load uniform ascent;

[0008] Step B, for different stress test points in different areas of the bucket, the stress data of the stress monitoring nodes under different operating conditions collected by the stress monitoring nodes are input into the long short-term memory neural network, and the overall stress data of each key area under different operating conditions of the bucket are predicted and output.

[0009] Preferably, based on the stress data of the stress monitoring nodes collected by the skip under different operating conditions, the state of the stress monitoring nodes at the corresponding positions is adjusted to working or sleeping.

[0010] Preferably, step A comprises:

[0011] Step A1, determining the stress distribution law of the key structure in the bucket under different connection modes, wherein the key structure includes a main hanging plate and a column;

[0012] Step A2, performing static analysis on the skip under no-load state, and obtaining the distribution law of various stresses caused by the skip's own gravity when the skip is hoisted;

[0013] Step A3, when the skip is fully loaded under the acceleration-deceleration operation condition, respectively analyzing and obtaining the stress distribution law of the skip under the acceleration lifting and emergency braking conditions when the skip works normally;

[0014] Step A4, under the condition of full load of the skip bucket, generate stress cloud map and displacement cloud map under the uniform speed operation of the skip bucket based on finite element simulation, analyze the stress distribution law when the skip bucket works at a uniform speed, and obtain the stress and displacement distribution results when the skip bucket is fully loaded and runs at a uniform speed;

[0015] Step A5. Based on the stress distribution of each key structure of the skip under different operating conditions, determine the key areas of the skip that need to be monitored.

[0016] Preferably, in step A1, it includes:

[0017] A1.1. The main suspension plate bears the total weight of the skip and the coal it carries. Conduct a simulation analysis on the main suspension plate to determine the stress distribution law of the main suspension plate.

[0018] A1.2. Analyze the columns in the skip. Based on the characteristics of various different connection methods of the columns, determine the stress distribution law of the columns.

[0019] A1.3. Analyze the bolts in the skip to determine the stress distribution law at each bolt connection.

[0020] Preferably, in step A2, conduct a static analysis of the skip under no-load conditions, calculate and obtain the stress distribution laws of the skip caused by its own gravity during hoisting, and obtain the stress and displacement distribution results of each key structure of the skip under no-load.

[0021] Preferably, in step A3, it includes:

[0022] A3.1. Determine the static load, dynamic load, impact load of coal on the skip, and impact load of acceleration during the acceleration and deceleration of the skip that the skip bears during operation.

[0023] A3.2. Based on the fact that the acceleration of the skip during hoisting and transportation shall not exceed a1 and the acceleration during braking of the skip shall not exceed a2, where a1 < a2, select various accelerations of different magnitudes between [a1, a2] for modeling analysis during the acceleration and deceleration of the fully loaded skip, and obtain stress nephograms and displacement nephograms under multiple accelerations.

[0024] A3.3. Based on the stress nephograms and displacement nephograms under multiple accelerations, analyze the stress distribution law of the fully loaded skip during acceleration or deceleration at different accelerations, and obtain the stress and displacement distribution results of the fully loaded skip during acceleration and deceleration.

[0025] Preferably, in step A4, it includes:

[0026] A4.1. Based on the speed range of the skip during uniform motion under actual working conditions, determine the speed range v1 - v2 of the fully loaded skip during uniform motion in the simulation software.

[0027] A4.2. Based on the speed range v1 - v2 of the skip during uniform motion, generate stress nephograms and displacement nephograms of the skip during uniform motion at different speeds, analyze the stress distribution law of the fully loaded skip during uniform motion at different speeds, and obtain the stress and displacement distribution results of the fully loaded skip during uniform motion.

[0028] Preferably, step B comprises:

[0029] Step B1, for different stress test points in the region, a relational model based on a data inversion algorithm is established using a long short-term memory network to obtain the transfer characteristics between different stress test points in the key region; some important positions in the key region of the bucket are selected to deploy stress monitoring nodes, and the stress monitoring nodes in the region are evenly arranged to obtain a variety of node deployment strategies for bucket stress monitoring; important positions include the connection of the key region, the midpoint of the key region and the vertex of the key region;

[0030] Step B2, input the stress data obtained by the stress monitoring node under the node deployment strategy into the long short-term memory network to obtain the regional stress prediction results under each node deployment strategy; compare the stress prediction results under different node deployment strategies, determine the optimal node deployment strategy, and determine the stress monitoring network of the bucket in the initial state.

[0031] Preferably, step B2 includes:

[0032] B2.1. Obtain the horizontal normal stress and vertical normal stress of the node groups to be tested in all areas based on the stress monitoring network;

[0033] B2.2. Under various operating conditions of the skip, the horizontal normal stress and vertical normal stress of the node group to be tested obtained in all regions are used as the input of the long short-term memory neural network, and the long short-term memory neural network outputs the predicted regional stress prediction results under each node deployment strategy.

[0034] Preferably, the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the methods described in the first aspect when executing the program.

[0035] Preferably, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0036] The beneficial effects achieved by the present invention are:

[0037] 1) For different stress test points in the region, a relational model of the data inversion algorithm was established based on the long short-term memory network (LSTM) to obtain the transfer characteristics between different stress test points in the key area, so as to invert the overall stress change in the key area of ​​the bucket by deploying a small number of key sensor nodes;

[0038] 2) In view of the problem that it is difficult to realize wired charging or frequent replacement of batteries in nodes in the harsh working environment underground, the present invention designs an array piezoelectric energy collection module and an energy management module, which can realize the collection and management of vibration energy of the bucket under different operating conditions to meet the energy consumption requirements of various loads on the bucket;

[0039] 3) Based on the energy collection module and the energy management module, a voltage frequency-based bucket operation speed perception strategy is designed to obtain the current operation speed of the bucket in real time. On this basis, a node adaptive working mode switching strategy is proposed. According to the stress distribution characteristics of the bucket under different operating conditions, the autonomous switching of the periodic working and sleep modes of nodes at different monitoring positions is realized to further reduce the energy consumption of the nodes. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0041] Figure 1 It is a schematic diagram of the steps of the present invention;

[0042] Figure 2 is a schematic diagram of an energy collection module in the present invention;

[0043] Figure 3 This is a schematic diagram of the structure of the energy management module in the present invention;

[0044] Figure 4 It is a flow chart of the skip running speed perception strategy based on voltage frequency in the present invention;

[0045] Figure 5 A schematic diagram of the deployment of the stress monitoring nodes of the bucket under conventional conditions in the present invention;

[0046] Figure 6 It is a schematic diagram of node deployment when the whole process bucket is running in the present invention;

[0047] Figure 7 Schematic diagram of a prediction model based on a long short-term memory network in the present invention;

[0048] Attached figures are as follows: 1. bucket; 2. bracket; 3. pressure vibrator; 4. piezoelectric cantilever beam; 5. mass block m1; 6. mass block m2; 7. mass block m3; 8. winch room control box; 9. wireless receiving box; 10. sheave; 11. wireless transmitting box; 12. end reinforcement rib; 13. column; 14. main suspension plate; 15. first longitudinal beam; 16. second longitudinal beam; 17. bolt. DETAILED DESCRIPTION

[0049] Embodiment 1

[0050] See also Figure 1 The present application discloses a system for self-powered lifting bucket stress monitoring based on data inversion, including multiple stress monitoring nodes, an energy collection module and an energy management module. The multiple stress monitoring nodes are installed on the bucket 1, and the energy collection module is connected to the energy management module; wherein the energy management module includes a bracket 2, a plurality of piezoelectric vibrators 3, a plurality of piezoelectric cantilever beams 4 and a plurality of mass blocks of different weights. The bucket 1, the bracket 2, the piezoelectric cantilever beam 4 and the mass block are fixedly connected in sequence, and the piezoelectric cantilever beam 4 is fixedly provided with a piezoelectric vibrator 3, the piezoelectric vibrator 3 is connected to the energy management module, and the energy management module is connected to the stress monitoring node.

[0051] Preferably, a method for self-powered hoisting bucket stress monitoring based on data inversion, using a system for self-powered hoisting bucket stress monitoring based on data inversion, performs:

[0052] Step A, using a finite element analysis method to obtain stress distribution in key areas of the bucket 1 under different working conditions, and determine the key areas that need to be monitored, wherein the different working conditions include no-load descent, full-load acceleration-deceleration ascent, and full-load uniform ascent;

[0053] Step B, for different stress test points in different areas of the bucket 1, the stress data of the stress monitoring nodes under different operating conditions collected by the stress monitoring nodes are input into the long short-term memory neural network, and the overall stress data of each key area under different operating conditions of the bucket 1 are predicted and output.

[0054] Preferably, based on the stress data of the stress monitoring nodes collected under different operating conditions of the skip 1 and the predicted overall stress data of each key area of ​​the skip 1, the state of the stress monitoring nodes at the corresponding positions is adjusted to working or sleeping.

[0055] Preferably, step A comprises:

[0056] Step A1, determining the stress distribution law of the key structure in the bucket 1 under different connection modes, wherein the key structure includes a main hanging plate and a column;

[0057] Step A2, performing static analysis on the skip 1 under no-load state, and obtaining the distribution law of various stresses caused by the self-gravity of the skip 1 when the skip 1 is hoisted;

[0058] Step A3, when the bucket 1 is fully loaded under the acceleration-deceleration operation condition, respectively analyzing and obtaining the stress distribution law of the bucket 1 under the acceleration and emergency braking conditions when the bucket 1 is working normally;

[0059] Step A4. Under the condition that the skip 1 is fully loaded, generate the stress nephogram and displacement nephogram of the skip 1 during uniform motion based on finite element simulation, analyze the stress distribution law of the skip 1 during uniform motion, and obtain the stress and displacement distribution results of the skip 1 during full-load uniform operation;

[0060] Step A5. Based on the stress distribution of each key structure of the skip 1 under different operating conditions, determine the key areas of the skip 1 that need to be monitored.

[0061] Preferably, in step A1, it includes:

[0062] A1.1. The main suspension plate bears the total weight of the skip 1 and the coal carried. Conduct a simulation analysis on the main suspension plate to determine the stress distribution law of the main suspension plate;

[0063] A1.2. Analyze the columns in the skip 1, and based on the characteristics of various different connection methods of the columns, determine the stress distribution law of the columns;

[0064] A1.3. Analyze the bolts in the skip 1 to determine the stress distribution law at each bolt connection.

[0065] Preferably, in step A2, conduct a static analysis on the skip 1 in the no-load state, calculate and obtain the stress distribution laws of each part caused by the self-weight of the skip 1 during hoisting, and obtain the stress and displacement distribution results of each key structure of the skip 1 in the no-load state.

[0066] Preferably, in step A3, it includes:

[0067] A3.1. Determine the static load, dynamic load, impact load of the coal on the skip 1, and impact load of the acceleration during the acceleration and deceleration of the skip 1 that the skip 1 bears during operation;

[0068] A3.2. Based on the fact that the acceleration of the skip 1 during hoisting and transportation shall not exceed a1 and the acceleration of the skip 1 during braking shall not exceed a2, where a1 < a2, select various accelerations of different magnitudes between [a1, a2] for modeling analysis during the full-load acceleration of the skip 1 to obtain the stress nephogram and displacement nephogram under multiple accelerations;

[0069] A3.3. Based on the stress nephogram and displacement nephogram under multiple accelerations, analyze the stress distribution law of the full-load skip 1 during acceleration or deceleration operation under different accelerations, and obtain the stress and displacement distribution results of the full-load skip 1 during acceleration and deceleration.

[0070] Preferably, in step A4, it includes:

[0071] A4.1. Based on the speed range of the skip 1 during uniform motion under actual working conditions, determine the speed range v1 - v2 of the skip 1 during full-load uniform motion in the simulation software;

[0072] A4.2. Based on the speed range v1-v2 of the bucket 1 in uniform motion, generate multiple sets of stress cloud maps and displacement cloud maps of the bucket 1 when it is running at different uniform speeds, analyze the stress distribution law of the bucket 1 in uniform motion when fully loaded at different speeds, and obtain the stress and displacement distribution results when the bucket 1 is fully loaded and running at a uniform speed.

[0073] Preferably, step B comprises:

[0074] Step B1, for different stress test points in the region, a relational model based on a data inversion algorithm is established using a long short-term memory network to obtain the transfer characteristics between different stress test points in the key region; some important positions in the key region of the bucket 1 are selected to deploy stress monitoring nodes, and the stress monitoring nodes in the region are evenly arranged to obtain a variety of node deployment strategies for bucket 1 stress monitoring; the important positions include the connection of the key region, the midpoint of the key region and the vertex of the key region;

[0075] Step B2, input the stress data obtained by the stress monitoring node under the node deployment strategy into the long short-term memory network to obtain the regional stress prediction results under each node deployment strategy; compare the stress prediction results under different node deployment strategies, determine the optimal node deployment strategy, and determine the stress monitoring network of bucket 1 in the initial state.

[0076] Preferably, step B2 includes:

[0077] B2.1. Obtain the horizontal normal stress and vertical normal stress of the node groups to be tested in all areas based on the stress monitoring network;

[0078] B2.2. Under various operating conditions of skip 1, the horizontal normal stress and vertical normal stress of the node group to be tested obtained in all regions are used as the input of the long short-term memory neural network, and the long short-term memory neural network outputs the predicted regional stress prediction results under each node deployment strategy.

[0079] In an embodiment of the present application, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods when executing the program.

[0080] In an embodiment of the present application, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0081] Embodiment 2

[0082] The first aspect of the present invention discloses a self-powered lifting bucket stress monitoring system based on data inversion, which includes an energy collection module, an energy management module and a node module.

[0083] like Figure 2 As shown, the energy collection module triggers energy collection by lifting the vibration of the bucket 1, and then obtains a periodic piezoelectric signal, and is characterized in that it includes a bracket 2, a piezoelectric vibrator 3 and a suspended piezoelectric component, wherein: the suspended piezoelectric component includes a piezoelectric cantilever beam 4, one end of the piezoelectric cantilever beam 4 is fixed on the bracket 2, and the other end of the piezoelectric cantilever beam 4 is suspended, and the suspended ends of the piezoelectric cantilever beam 4 are respectively provided with mass blocks of different weights, and the mass blocks of different weights form an array-type piezoelectric energy collection device; when the bucket 1 starts to be lifted, the bucket 1 body generates vibrations of different frequencies. At this time, based on the piezoelectric effect, the array-type piezoelectric energy collection device can convert vibration energy into electrical energy to realize the supplement of stress monitoring node energy; the bracket 2 is arranged on the bucket 1, and the piezoelectric cantilever beam 4 is always in a suspended state during the operation of the lifting bucket 1;

[0084] As a further optimization scheme of the present invention, Figure 3-4 As shown, the energy management module adopts the LTC3331 chip to rectify, filter, boost / step down the collected electrical signals of the array piezoelectric energy collection device, store the excess energy in the node battery, and can sense the current operating speed of the lifting bucket 1 by periodically monitoring the frequency of the power generation voltage of the array piezoelectric energy collection device.

[0085] Specifically, when the collected energy is high voltage, it is stepped down and supplies energy to the node, while storing excess energy in the node battery; when the collected energy is low voltage, it automatically switches to the lithium battery to supply energy to the load node;

[0086] Specifically, the simple sine wave voltage generated by the piezoelectric vibrator is first converted into a square wave voltage by using the comparator of the microprocessor; then the high and low pulse periods of the square wave voltage in unit time are detected; finally, the frequency of the bucket to which the piezoelectric vibrator is subjected is obtained by calculation; and the excitation frequency of the bucket vibration to which the piezoelectric vibrator is subjected is related to the running speed of the bucket, so the frequency of the voltage generated by the piezoelectric vibrator is related to the running speed of the bucket, thereby obtaining the current running speed of the lifting bucket.

[0087] Specifically, part of the AC signal generated by the energy collection device enters the energy management module, which is mainly used to power the wireless sensor nodes, microprocessors and signal pre-processing; the other part converts the generated sinusoidal AC signal into a square wave signal that can be detected by the microprocessor through signal pre-processing, that is, the frequency value f of the current AC signal is obtained by detecting the period of high and low pulses in unit time, and then the running speed v of the bucket is obtained according to the relationship between the power generation frequency and the bucket running speed, which is expressed as: v=k×f, where k is the conversion coefficient of the device.

[0088] The LTC3331 chip is used to rectify, filter, boost / step down the collected electrical signals, store the excess energy in the node battery, and sense the current operating speed of the lifting bucket by periodically monitoring the frequency of the piezoelectric vibrator's generated voltage;

[0089] As a further optimization scheme of the present invention, the node module includes: step A, using a finite element analysis method to obtain stress distribution conditions in key areas of the bucket 1 under different working conditions, and determining key areas that need to be monitored; the different working conditions include no-load descent, full-load acceleration-deceleration ascent, and full-load uniform ascent;

[0090] Step B, for different stress test points in the area, use the long short-term memory network to establish a relationship model based on the data inversion algorithm to obtain the transfer characteristics between different stress test points in the key area; determine the deployment strategy of sensors in the key area based on the transfer characteristics; input the data set including the stress data of the stress monitoring nodes collected under different operating conditions into the long short-term memory neural network, and use the long short-term memory neural network to predict the overall stress data of each key area under different working conditions of the bucket 1, so as to realize the stress data prediction of the whole process in the key area;

[0091] The node adaptive working mode switching strategy includes: based on the data set of stress data of the stress monitoring node of bucket 1 under different operating conditions, adjusting the state of the stress monitoring node at the corresponding position to working or sleeping, realizing the autonomous switching of the periodic working and sleeping modes of the stress monitoring nodes at different monitoring positions, and further reducing the energy consumption of the stress monitoring nodes.

[0092] A certain type of 50t side-loading skip was selected, and the detection nodes were deployed in important areas of the skip based on the self-powered lifting skip stress monitoring method based on data inversion.

[0093] like Figure 1 As shown, a self-powered lifting bucket stress monitoring method based on data inversion includes:

[0094] A. Use finite element analysis method to obtain the stress distribution of key areas of bucket 1 under different working conditions, and determine the key areas that need to be monitored; different working conditions include no-load descent, full-load acceleration-deceleration ascent, and full-load uniform ascent;

[0095] B. For different stress test points in the region, a relational model based on a data inversion algorithm is established using a long short-term memory network to obtain the transfer characteristics between different stress test points in the key region; the deployment strategy of sensors in the key region is determined based on the transfer characteristics; the data set including the stress data of the stress monitoring nodes collected under different operating conditions is input into the long short-term memory neural network, and the long short-term memory neural network is used to predict the overall stress data of each key region under different operating conditions of Jidou 1, so as to realize the stress data prediction of the whole process in the key region;

[0096] C. According to the vibration frequency f of the bucket 1 under different operating conditions i (f i ∈f1, f2 and f3, f1, f2 and f3 represent the vibration frequency of no-load, the vibration frequency of full-load acceleration-deceleration and the vibration frequency of full-load constant speed respectively), an array piezoelectric vibration energy collection module and energy management module are designed, based on the piezoelectric effect to realize the collection and management of the vibration energy of the bucket 1 under different working conditions;

[0097] D. On the basis of the energy collection module and the energy management module, the current running speed of the bucket 1 is obtained by using the voltage frequency-based operating speed perception strategy of the bucket 1. The node adaptive working mode switching strategy includes: based on the data set of stress data of the stress monitoring node of the bucket 1 under different operating conditions, the state of the stress monitoring node at the corresponding position is adjusted to work or sleep, and the periodic working and sleep modes of the stress monitoring nodes at different monitoring positions are autonomously switched, and the energy consumption of the stress monitoring nodes is further reduced. According to the stress distribution characteristics of the stress data of the stress monitoring nodes under different operating conditions collected by the bucket 1, the periodic working and sleep modes of the stress monitoring nodes at different monitoring positions are autonomously switched, and the energy consumption of the stress monitoring nodes is further reduced.

[0098] The step A comprises:

[0099] A1, determine the stress distribution law of the key structure in the bucket 1 under different connection modes, wherein the key structure includes the main hanging plate and the column;

[0100] A2, the statics analysis of the skip 1 under the no-load state is performed, i.e., the stress distribution law caused by the self-gravity of the skip 1 when hoisting is calculated by simulation software;

[0101] A3, the actual working of the skip 1 goes through three stages, i.e., acceleration, uniform speed and deceleration. First, when the skip 1 is fully loaded under the acceleration-deceleration operation condition, the stress distribution law of the skip 1 under the acceleration lifting and emergency braking conditions when the skip 1 is working normally is analyzed respectively;

[0102] A4. Under the condition of full load of bucket 1, the stress cloud map and displacement cloud map under uniform speed operation are generated based on finite element simulation, the stress distribution law of bucket 1 under uniform speed operation is analyzed, and the stress and displacement distribution results of bucket 1 under full load and uniform speed operation are obtained;

[0103] A5. Based on the stress distribution of each key structure of skip 1 under different operating conditions, determine the key areas of skip 1 that need to be monitored.

[0104] Further, in step A, taking a certain type of 50-ton side-loading bucket as an example, the stress cloud diagram of each key structure is obtained based on finite element modeling analysis, and the stress distribution law of the key structure in the large bucket under various connection modes is analyzed. The key structure mainly includes the main hanging plate 14, the column 13, etc., as follows:

[0105] A1.1. The main hanging plate 14 bears the entire weight of the skip body and the coal carried. First, a simulation analysis is performed on the main hanging plate 14 to determine the stress distribution law of the main hanging plate;

[0106] A1.2. Analyze the columns 13 in the bucket and determine the stress distribution law of each structure based on the characteristics of various connection modes of the columns 13;

[0107] A1.3. Finally, the bolts 17 commonly used in the skip are analyzed to determine the stress distribution pattern at each bolt connection in the skip.

[0108] Among them, analysis shows that the maximum stress of the bucket is concentrated on the upper parts of both sides of the main hanger plate 14 in the upper plate suspension device and on the columns 13 in the bucket chassis. The stress distribution of the bolt group 17 is relatively large, and mostly appears at the root of the bolt 17.

[0109] Furthermore, in step A2, a static analysis is performed on the skip in an unloaded state, that is, the stress distribution laws caused by the self-gravity of the skip 1 during hoisting are simulated and calculated using simulation software to obtain the stress and displacement distribution results of each key structure when the skip is unloaded.

[0110] Among them, the stress range of the entire bucket when unloaded is 3.924×10-4~141.99MPa, and the maximum stress occurs in the upper plate suspension device.

[0111] Furthermore, in step A3, when the skip is actually working, it experiences three stages, namely acceleration, constant speed, and deceleration. First, considering the acceleration-deceleration operating conditions, when the skip is fully loaded, the stress distribution laws during normal acceleration lifting and emergency braking are analyzed respectively to obtain the stress and displacement distribution results of the fully loaded skip during acceleration-deceleration, as follows:

[0112] A3.1. Determine the static load, dynamic load, impact load of coal on the skip 1, and impact load of acceleration during acceleration-deceleration of the skip 1 that the skip 1 bears during operation;

[0113] A3.2. Based on the acceleration of the skip 1 during lifting and transportation not exceeding a1 and the acceleration of the skip 1 during braking not exceeding a2, where a1 < a2, between the above two acceleration thresholds, multiple accelerations of different magnitudes are selected for modeling analysis during the full-load acceleration of the skip 1 to obtain stress nephograms and displacement nephograms under multiple accelerations;

[0114] A3.3. Based on the stress nephograms and displacement nephograms under multiple accelerations, analyze the stress distribution laws during the acceleration or deceleration operation of the fully loaded skip 1 under different accelerations to obtain the stress and displacement distribution results of the fully loaded skip 1 during acceleration-deceleration.

[0115] Among them, based on different accelerations, the statistical results of the stress analysis of multiple finite element analyses of the skip are as follows:

[0116] The overall stress distribution of the skip is mainly concentrated in the upper suspension device (main hanging plate 14, first longitudinal beam 5, and first longitudinal beam 6), the middle and lower parts of the skip box, especially the lower half of the lower box of the skip is particularly prominent; the maximum stress position appears at the right end of the stiffener 12 at the front end of the box body; the displacement of the upper half of the skip body is small, and the displacement of the lower box of the skip is significant; the maximum displacement appears below the stiffener 12 at the lower end of the skip box and about one-third of the distance from the unloading port of the box body.

[0117] Furthermore, in step A4, under the condition that the skip 1 is fully loaded, based on the finite element simulation, stress nephograms and displacement nephograms during its constant-speed operation are generated, and the stress distribution law during the constant-speed operation of the skip is analyzed to obtain the stress and displacement distribution results during the full-load constant-speed operation of the skip, as follows:

[0118] A4.1. Based on the speed range during the constant-speed operation of the skip under actual working conditions, determine the speed range of 5 m / s - 15 m / s for the full-load constant-speed movement of the skip in the simulation software;

[0119] A4.2. Based on the speed range of 5 m / s - 15 m / s during the constant-speed movement of the skip, generate stress nephograms and displacement nephograms of the skip during constant-speed operation at different speeds, analyze the stress distribution law during the constant-speed movement of the fully loaded skip at different speeds, and obtain the stress and displacement distribution results during the full-load constant-speed operation of the skip.

[0120] Among them, from the finite element stress analysis results of multiple groups of skips at full load and uniform speed, it can be seen that the maximum stress of the skip is concentrated on the main hanger plate 14 in the upper plate suspension device, and the maximum deformation is mostly distributed in the middle area of ​​the lower middle part of the skip lower box.

[0121] Furthermore, in step A5, based on the stress distribution of each key part of the skip under different working conditions, the key areas that need to be monitored are determined to achieve real-time monitoring of the stress in the key areas.

[0122] Among them, in the no-load state, the stress is concentrated on the upper plate suspension device of the bucket (the main suspension plate 14 and the first longitudinal beam 5 and the first longitudinal beam 6). Therefore, the upper plate suspension device is the key monitoring area S1 under this working condition;

[0123] When the bucket is fully loaded and moves at a uniform speed, the maximum stress of the bucket is concentrated in the middle upper part of the upper plate suspension device, the first longitudinal beams 5 and 6 on both sides connected to the main hanging plate 14, and the column 13 in the bucket chassis, that is, area S1, area S2 and area S3, and the maximum deformation is mostly distributed in the middle of the middle and lower part of the bucket body, that is, area S4;

[0124] When the bucket is fully loaded and accelerates and decelerates, the maximum stress position is concentrated at the right end of the end reinforcement rib 12 of the box front, that is, area S5; the maximum deformation mostly occurs below the end reinforcement rib 12 of the lower part of the bucket box, about one-third of the distance from the unloading port of the box, that is, area S6.

[0125] Further, in step B, the details are as follows:

[0126] B1. For different stress test points in the region, a relational model based on a data inversion algorithm is established using a long short-term memory network to obtain the transfer characteristics between different stress test points in the key region; stress monitoring nodes are deployed at some important positions in the key region of the bucket 1, and the stress monitoring nodes in the region are evenly arranged at intervals of d, 2d, 3d, 4d, ..., so as to determine a variety of node deployment strategies for bucket 1 stress monitoring; important positions include the connection points of the key region, the midpoints of the key region, and the vertices of the key region;

[0127] B2. Input the stress data obtained by the stress monitoring nodes under the node deployment strategy into the long short-term memory network to obtain the regional stress prediction results under each node deployment strategy; compare the stress prediction results under different node deployment strategies, determine the optimal node deployment strategy, and determine the stress monitoring network of bucket 1 in the initial state.

[0128] Among them, the bucket is simplified into a cuboid space deployment model, stress monitoring node 1 and stress monitoring node 2 are arranged on the central axis of the lower end face of the cuboid, and the average straight-line distance between stress monitoring node 1 and stress monitoring node 2 is d1, which serves as two monitoring node groups on the first longitudinal beam 5 and the first longitudinal beam 6 on both sides connected to the main hanger 14 to realize stress monitoring of the key monitoring area S2; stress monitoring node 3 is arranged on the central axis of the left side of the cuboid at a distance of d2 from the lower end face, and stress monitoring node 4 is arranged on the central axis of the right side of the cuboid at the same distance from the stress monitoring node The corresponding position of point 3 is used as two monitoring node groups 13 on the symmetrical columns in the bucket to realize stress monitoring of the key monitoring area S3; the stress monitoring node 5 is arranged on the central axis of the front view of the rectangular parallelepiped at a distance of d4 from the upper end face to realize stress monitoring of the key monitoring area S4, and the stress monitoring node 6 is arranged on the central axis of the front view of the rectangular parallelepiped at a distance of d3 from the lower end face to realize stress monitoring of the key monitoring areas S5 and S6. The stress monitoring node 7 is arranged at the intersection of the central axis of the front view of the rectangular parallelepiped and the upper end face to realize stress monitoring of area S1. Here, it should be noted that for the convenience of representation, only one node is arranged in the key monitoring area, and this node can represent the node group in the area.

[0129] As a further optimization scheme of the present invention, the specific implementation steps in step B2 are as follows:

[0130] B2.1. Obtain the horizontal normal stress and vertical normal stress of the node groups to be tested in all areas based on the stress monitoring network;

[0131] B2.2. Under various operating conditions of bucket 1, the horizontal normal stress and vertical normal stress of the node group to be tested obtained in all regions are used as the input of the long short-term memory neural network, and the regional stress prediction results under each node deployment strategy are obtained through the LSTM layer, attention mechanism and full connection layer of the long short-term memory neural network in sequence. Specifically, first, a layer of LSTM is used to make full use of the timing characteristics in each region, then the attention mechanism is used to extract the input features that play a key role in stress data prediction, and finally, the full connection layer is used to output the prediction results. Repeat the above iterative process to achieve real-time prediction of complete stress data in key monitoring areas 1, area 2, ... area m.

[0132] B2.3, when the bucket is fully loaded and accelerating upward (decelerating upward), the positive stress data sets of the stress monitoring node groups to be tested in the horizontal and vertical directions obtained from regions S1, S2, S3, S5, S6 and S7 are used as the input of the long short-term memory neural network, and are sequentially passed through the LSTM, attention mechanism and fully connected layer of the network. First, a layer of LSTM is used to fully utilize the timing characteristics in each region, and then the attention mechanism is used to extract the input features that play a key role in stress data prediction. Finally, the fully connected layer is used to output the prediction results. Repeat the above iterative process to achieve real-time prediction of complete stress data in the key monitoring areas S1, S2, S3, S5, S6 and S7;

[0133] B2.4, when the bucket rises at a uniform speed, the positive stress data sets of the node groups to be tested in the horizontal and vertical directions obtained from regions S1, S2, S3, S4 and S7 are used as the input of the long short-term memory neural network, and are sequentially passed through the LSTM, attention mechanism and fully connected layer of the network. First, a layer of LSTM is used to fully utilize the time series characteristics in each region, and then the attention mechanism is used to extract the input features that play a key role in stress data prediction. Finally, the fully connected layer is used to output the prediction results. Repeat the above iterative process to achieve real-time prediction of complete stress data in the key monitoring areas S1, S2, S3, S4 and S7;

[0134] B2.5. Based on the above steps B2.2-B2.4, output the stress distribution results of the whole process in the key monitoring area of ​​the skip.

[0135] Further, in step C, according to the vibration frequency f of the bucket under different working conditions i (f1, f2 and f3 represent the vibration frequencies at no load, full load acceleration and deceleration, and full load constant speed, respectively). An array piezoelectric vibration energy collection module and energy management module are designed to realize the collection and management of the bucket vibration energy under different working conditions based on the piezoelectric effect.

[0136] Furthermore, in step D, based on the energy collection module and the energy management module, a voltage frequency-based bucket operation speed perception strategy is used to obtain the current operation speed of the bucket, so as to realize autonomous switching of periodic operation and sleep modes of nodes at different monitoring positions according to the stress distribution characteristics of the bucket under different working conditions, thereby further reducing the energy consumption of the nodes.

[0137] Figure 5 This is a schematic diagram of node deployment for monitoring the stress state of the skip under normal conditions, wherein each stress sensor is evenly arranged on the outer wall of the skip 1 at a spacing d.

[0138] Figure 6The diagram shows the node deployment diagram of the whole process of the present invention when the bucket is running, including:

[0139] Schematic diagram of node deployment when the bucket is running without load, where, when the bucket is running without load, nodes 1, 2, and 7 are working normally, and the remaining nodes are in a dormant state; (yellow nodes indicate that they are working normally, and white nodes indicate that they are in a dormant state);

[0140] Schematic diagram of node deployment when the bucket is fully loaded and moving at a constant speed. When the bucket is fully loaded and moving at a constant speed, nodes 1, 2, 3, 4, 5, and 7 are working normally, and node 6 is in a dormant state.

[0141] Schematic diagram of node deployment when the bucket is fully loaded and accelerating and decelerating. When the bucket is fully loaded and accelerating and decelerating, nodes 1, 2, 3, 4, 6, and 7 work normally, and node 5 is in a dormant state.

[0142] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0143] Those skilled in the art will readily appreciate other embodiments of the invention after considering the specification and practicing the invention invented herein. This application is intended to cover any variations, uses or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art that are not invented by the present invention. The specification and examples are intended to be exemplary only, and the true scope and spirit of the invention are indicated by the claims.

[0144] The above specific implementation methods further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above are only specific implementation methods of the present application and are not used to limit the protection scope of the present application. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present application should be included in the protection scope of the present application.

Claims

1. A self-powered lifting bucket stress monitoring system based on data inversion, characterized in that: The invention comprises a plurality of stress monitoring nodes, an energy collection module and an energy management module. The plurality of stress monitoring nodes are installed on a bucket (1). The energy collection module is connected to the energy management module. The energy management module comprises a bracket (2), a plurality of piezoelectric vibrators (3), a plurality of piezoelectric cantilever beams (4) and a plurality of mass blocks of different weights. The bucket (1), the bracket (2), the piezoelectric cantilever beams (4) and the mass blocks are fixedly connected in sequence. The piezoelectric cantilever beams (4) are fixedly provided with piezoelectric vibrators (3). The piezoelectric vibrators (3) are connected to the energy management module. The energy management module is connected to the stress monitoring nodes.

2. A method for monitoring stress of a self-powered lifting bucket based on data inversion, characterized in that: Using the self-powered lifting bucket stress monitoring system based on data inversion as described in claim 1, executing: Step A, using a finite element analysis method to obtain stress distribution conditions in key areas of the bucket (1) under different working conditions, and determine key areas that need to be monitored, wherein the different working conditions include no-load descent, full-load acceleration-deceleration ascent, and full-load uniform ascent; Step B, for different stress test points in different areas of the bucket (1), input the stress data of the stress monitoring node under different operating conditions collected by the stress monitoring node into the long short-term memory neural network, and predict and output the overall stress data of each key area of ​​the bucket (1) under different operating conditions.

3. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 2, characterized in that: Based on the stress data of the stress monitoring nodes collected by the bucket (1) under different operating conditions, the state of the stress monitoring nodes at the corresponding positions is adjusted to work or sleep.

4. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 2, characterized in that: The step A comprises: Step A1, determining the stress distribution law of the key structure in the bucket (1) under different connection modes, wherein the key structure includes the main hanging plate and the column; Step A2, performing a static analysis on the bucket (1) in an unloaded state, and obtaining the distribution rules of various stresses caused by the bucket (1)'s own gravity when the bucket (1) is hoisted; Step A3, when the bucket (1) is fully loaded under the acceleration-deceleration operation condition, respectively analyzing and obtaining the stress distribution law of the bucket (1) under the acceleration and lifting conditions and the emergency braking conditions during normal operation of the bucket (1); Step A4, generating a stress cloud map and a displacement cloud map of the bucket (1) under uniform speed operation based on finite element simulation under the condition that the bucket (1) is fully loaded, analyzing the stress distribution law of the bucket (1) when it is working at a uniform speed, and obtaining the stress and displacement distribution results of the bucket (1) when it is fully loaded and running at a uniform speed; Step A5: based on the stress distribution of each key structure of the skip (1) under different operating conditions, determine the key area of ​​the skip (1) that needs to be monitored.

5. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 4, characterized in that: Step A1 includes: A1.

1. The main hanging plate bears the entire weight of the skip (1) and the coal carried. A simulation analysis is performed on the main hanging plate to determine the stress distribution law of the main hanging plate. A1.

2. Analyze the columns in the bucket (1) and determine the stress distribution law of the columns based on the characteristics of various connection methods of the columns; A1.

3. Analyze the bolts in the bucket (1) and determine the stress distribution pattern at each bolt connection.

6. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 4, characterized in that: In step A2, a static analysis is performed on the bucket (1) in an unloaded state, and the stress distribution laws caused by the bucket (1)'s own gravity when the bucket (1) is hoisted are calculated to obtain the stress and displacement distribution results of each key structure of the bucket (1) when it is unloaded.

7. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 4, characterized in that: Step A3 includes: A3.

1. Determine the static load and dynamic load borne by the bucket (1) during operation, the impact load of the coal on the bucket (1), and the impact load of the acceleration of the bucket (1) during acceleration and deceleration; A3.

2. The acceleration during lifting and transportation based on the bucket (1) shall not exceed a1 The acceleration of the bucket (1) during braking shall not exceed a2 ,in, a1 <a2 ,exist[ a1 , a2 ] select a variety of accelerations of different sizes to perform modeling analysis on the bucket (1) when fully loaded with acceleration, and obtain stress cloud diagrams and displacement cloud diagrams under multiple groups of accelerations; A3.

3. Based on multiple sets of stress cloud diagrams and displacement cloud diagrams under acceleration, the stress distribution law of the fully loaded bucket (1) during acceleration or deceleration under different accelerations is analyzed, and the stress and displacement distribution results of the fully loaded bucket (1) during acceleration and deceleration are obtained.

8. The method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 4, characterized in that: Step A4 includes: A4.

1. Based on the speed range of the bucket (1) when it is running at a constant speed under actual working conditions, determine the speed range of the bucket (1) when it is fully loaded and moving at a constant speed in the simulation software. v1 - v2 ; A4.

2. Based on the range of speed of the bucket (1) when in uniform motion v1 - v2 , generate multiple groups of stress cloud maps and displacement cloud maps when the bucket (1) runs at different speeds, analyze the stress distribution law of the bucket (1) running at a uniform speed when fully loaded at different speeds, and obtain the stress and displacement distribution results when the bucket (1) runs at a uniform speed when fully loaded.

9. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 2, characterized in that: Step B includes: Step B1, for different stress test points in the region, a relational model based on a data inversion algorithm is established using a long short-term memory network to obtain the transfer characteristics between different stress test points in the key region; stress monitoring nodes are deployed at some important positions in the key region on the bucket (1), and the stress monitoring nodes in the region are evenly arranged, thereby obtaining a variety of node deployment strategies for stress monitoring of the bucket (1); the important positions include the connection points of the key region, the midpoints of the key region, and the vertices of the key region; Step B2, input the stress data obtained by the stress monitoring node under the node deployment strategy into the long short-term memory network to obtain the regional stress prediction results under each node deployment strategy; compare the stress prediction results under different node deployment strategies, determine the optimal node deployment strategy, and determine the stress monitoring network of the bucket (1) in the initial state.

10. A method for monitoring stress of a self-powered lifting bucket based on data inversion according to claim 9, characterized in that: Step B2 includes: B2.

1. Obtain the horizontal normal stress and vertical normal stress of the node groups to be tested in all areas based on the stress monitoring network; B2.

2. Under various operating conditions of the bucket (1), the horizontal normal stress and vertical normal stress of the node group to be tested obtained in all regions are used as the input of the long short-term memory neural network, and the long short-term memory neural network outputs the regional stress prediction results under each node deployment strategy.

Images