Anti-collision control system and method for bucket-wheel stacker reclaimer

Abstract

The invention relates to the field of bucket-wheel stacker-reclaimers, and particularly discloses an anti-collision control system and method for a bucket-wheel stacker-reclaimer. After an obstacle is detected, a corresponding speed reduction strategy is triggered based on the position of the obstacle, and meanwhile, the dust concentration and air humidity in an operation environment are monitored in real time; and performing data analysis on the dust concentration and air humidity data by adopting a time sequence analysis technology based on deep learning so as to extract time sequence fluctuation characteristics of the dust concentration and the air humidity, and further capturing cooperative change between the dust concentration and the air humidity by mining implicit information association between the dust concentration and the air humidity. And the deceleration strategy is corrected on the basis, so that the motion of the bucket-wheel stacker-reclaimer is controlled according to the corrected deceleration. According to the method, the comprehensive influence of multi-dimensional environmental factors on the laser detection performance is comprehensively considered, so that the anti-collision performance of the bucket-wheel stacker-reclaimer in a complex environment can be effectively improved, the collision risk is reduced, and the operation safety is ensured.

Description

Technical Field

[0001] The present application relates to the technical field of bucket-wheel stacker reclaimers, and particularly relates to an anti-collision control system and method for a bucket-wheel stacker reclaimer. Background Art

[0002] A bucket-wheel stacker reclaimer (BSR) is an important loading and unloading device widely used in large dry bulk yards, capable of efficiently performing material stacking and extraction operations. With the development of automation and unmanned technologies, the application of bucket-wheel stacker reclaimers in modern ports, mines, and industrial production is becoming increasingly widespread. However, due to its complex and variable working environment, especially in an environment with high dust concentration and large humidity changes, how to ensure the safe operation of the bucket-wheel stacker reclaimer is a major challenge currently faced.

[0003] In this regard, the invention patent with the publication number CN115576360A discloses an anti-collision control method for a bucket wheel machine, which uses a laser detection system to obtain road condition information, initially adjusts the running speed of the bucket wheel machine according to the road condition information, and corrects the initial adjustment according to the dust particle content and air humidity data to compensate for the influence of environmental factors on the laser detection accuracy. Furthermore, after a preset time T1, the position of the obstacle is detected again, the position and movement state of the obstacle are judged, and the running state is adjusted again until the obstacle is passed.

[0004] However, although the prior art considers the influence of dust concentration and air humidity on laser detection and improves the running safety of the bucket wheel machine to a certain extent, its correction strategy is relatively simple and mainly relies on a preset correction coefficient to linearly correct the speed of the bucket wheel machine. In the actual working environment, the dust concentration and air humidity often show complex dynamic change characteristics, and the two have complex interactive effects on the performance of laser detection. The traditional preset correction coefficient usually cannot accurately reflect this complex interactive effect, resulting in poor correction effect.

[0005] Therefore, an optimized anti-collision control system and method for a bucket-wheel stacker reclaimer are expected. Summary of the Invention

[0006] Embodiments of the present application aim to at least solve one of the technical problems existing in the prior art, and provide an anti-collision control system and method for a bucket wheel stacker-reclaimer. After detecting an obstacle, it triggers a corresponding deceleration strategy based on the position of the obstacle, and simultaneously monitors the dust concentration and air humidity in the working environment in real time. It uses time series analysis technology based on deep learning to analyze the data of dust concentration and air humidity, so as to extract the time series fluctuation characteristics of dust concentration and air humidity. Furthermore, by mining the implicit information correlation between the two, it captures the co-variation between dust concentration and air humidity, and based on this, modifies the deceleration strategy to control the movement of the bucket wheel stacker-reclaimer according to the modified deceleration. This method can effectively improve the anti-collision performance of the bucket wheel stacker-reclaimer in a complex environment, reduce the collision risk, and ensure the operation safety by comprehensively considering the comprehensive influence of multi-dimensional environmental factors on the laser detection performance.

[0007] On the one hand, the present application provides an anti-collision control method for a bucket wheel stacker-reclaimer, which includes: using a laser detection system to scan the road conditions within a 270° range around; recording the position when an obstacle is detected for the first time, and obtaining the data of the bucket wheel machine speed, acceleration, dust particle content and air humidity; using an adjustment module to initially adjust the operation state of the bucket wheel machine, and correcting the adjustment based on the environmental data; after an interval time T1, detecting the position and movement state of the obstacle again, and adjusting the operation state of the bucket wheel machine again according to the detection results until the bucket wheel machine completely passes the obstacle. Using the adjustment module to initially adjust the operation state of the bucket wheel machine and correcting the adjustment based on the environmental data includes:

[0008] If the obstacle is in the front area, trigger an emergency deceleration strategy; if the obstacle is not in the front area, trigger a secondary deceleration strategy and alarm;

[0009] Obtain the time queue of real-time dust concentration values and the time queue of real-time air humidity values;

[0010] Extract the time series fluctuation characteristics of dust concentration and the time series fluctuation characteristics of air humidity from the time queue of real-time dust concentration values and the time queue of real-time air humidity values;

[0011] Conduct an interactive analysis between environmental parameters based on implicit information correlation on the time series fluctuation characteristics of dust concentration and the time series fluctuation characteristics of air humidity to obtain the dust concentration-air humidity time series co-variation characteristics;

[0012] Fuse the time series fluctuation characteristics of dust concentration, the time series fluctuation characteristics of air humidity and the dust concentration-air humidity time series co-variation characteristics to obtain multi-dimensional time series characteristics of environmental parameters;

[0013] Based on the multi-dimensional time series characteristics of environmental parameters, modify the deceleration strategy to obtain a modified deceleration.

[0014] Optionally, the deceleration of the emergency deceleration strategy and the deceleration of the secondary deceleration strategy are the first deceleration and the second deceleration respectively, and the first deceleration is greater than the second deceleration.

[0015] Optionally, extracting the dust concentration time series fluctuation feature and the air humidity time series fluctuation feature from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value includes:

[0016] Using a time series encoder based on the GRU model to perform time series feature extraction on the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value respectively to obtain the dust concentration time series fluctuation feature and the air humidity time series fluctuation feature.

[0017] Optionally, performing an interactive analysis between environmental parameters based on implicit information association on the dust concentration time series fluctuation feature and the air humidity time series fluctuation feature to obtain the dust concentration-air humidity time series collaborative feature, including:

[0018] Performing a feature principal component analysis on the air humidity time series fluctuation feature to obtain a set of principal component components of the air humidity time series fluctuation feature;

[0019] Performing an implicit association analysis between the dust concentration time series fluctuation feature and each principal component component of the air humidity time series fluctuation feature in the set of principal component components of the air humidity time series fluctuation feature to construct a {dust concentration time series fluctuation deep implicit feature, anchored air humidity time series fluctuation principal component deep implicit feature} feature pair;

[0020] Performing a fine-grained interactive analysis on the {dust concentration time series fluctuation deep implicit feature, anchored air humidity time series fluctuation principal component deep implicit feature} feature pair to obtain the dust concentration-air humidity time series collaborative feature.

[0021] Optionally, performing an implicit association analysis between the dust concentration time series fluctuation feature and each principal component component of the air humidity time series fluctuation feature in the set of principal component components of the air humidity time series fluctuation feature to construct a {dust concentration time series fluctuation deep implicit feature, anchored air humidity time series fluctuation principal component deep implicit feature} feature pair, including:

[0022] Performing an implicit feature extraction based on point convolution coding on the dust concentration time series fluctuation feature to obtain a dust concentration time series fluctuation deep implicit feature;

[0023] Performing an implicit feature extraction based on point convolution coding on each principal component component of the air humidity time series fluctuation feature in the set of principal component components of the air humidity time series fluctuation feature to obtain a set of air humidity time series fluctuation principal component deep implicit features;

[0024] Input the set of the deep implicit features of the temporal fluctuations of the dust concentration and the deep principal component implicit features of the temporal fluctuations of the air humidity into the implicit key clue anchoring network to obtain the {deep implicit features of the temporal fluctuations of the dust concentration, anchored deep principal component implicit features of the temporal fluctuations of the air humidity} feature pair.

[0025] Optionally, input the set of the deep implicit features of the temporal fluctuations of the dust concentration and the deep principal component implicit features of the temporal fluctuations of the air humidity into the implicit key clue anchoring network to obtain the {deep implicit features of the temporal fluctuations of the dust concentration, anchored deep principal component implicit features of the temporal fluctuations of the air humidity} feature pair, including:

[0026] Calculate the feature correlation factors between the deep implicit features of the temporal fluctuations of the dust concentration and each of the deep principal component implicit features of the temporal fluctuations of the air humidity in the set, and select the deep principal component implicit feature of the temporal fluctuations of the air humidity corresponding to the maximum feature correlation factor as the anchored deep principal component implicit feature of the temporal fluctuations of the air humidity to construct the {deep implicit features of the temporal fluctuations of the dust concentration, anchored deep principal component implicit features of the temporal fluctuations of the air humidity} feature pair.

[0027] Optionally, perform fine-grained interaction analysis on the {deep implicit features of the temporal fluctuations of the dust concentration, anchored deep principal component implicit features of the temporal fluctuations of the air humidity} feature pair to obtain the dust concentration-air humidity temporal coordination feature, including:

[0028] Perform feature power-law distribution alignment constraint on the deep implicit features of the temporal fluctuations of the dust concentration and the anchored deep principal component implicit features of the temporal fluctuations of the air humidity to obtain the {optimized deep implicit features of the temporal fluctuations of the dust concentration, optimized anchored deep principal component implicit features of the temporal fluctuations of the air humidity} feature pair;

[0029] Input the {optimized deep implicit features of the temporal fluctuations of the dust concentration, optimized anchored deep principal component implicit features of the temporal fluctuations of the air humidity} feature pair into the feature fine-grained interaction network to obtain the dust concentration-air humidity temporal coordination feature.

[0030] Optionally, based on the multi-dimensional temporal features of the environmental parameters, correct the deceleration strategy to obtain the corrected deceleration, including:

[0031] Input the multi-dimensional temporal features of the environmental parameters into the correction coefficient estimation module based on the decoder to obtain the environmental impact correction coefficient;

[0032] Multiply the environmental impact correction coefficient by the first deceleration or the second deceleration to obtain the corrected deceleration.

[0033] On the other hand, the present application provides an anti-collision control system for a bucket wheel stacker-reclaimer, including:

[0034] An obstacle detection module, configured to use a laser detection system to scan the road conditions information within a range of 270° around, trigger an emergency deceleration strategy if an obstacle is in the front area, and trigger a secondary deceleration strategy and alarm if the obstacle is not in the front area;

[0035] A real-time data acquisition module, configured to acquire the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value;

[0036] A time series feature extraction module, configured to extract the dust concentration time series fluctuation feature and the air humidity time series fluctuation feature from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value;

[0037] An environmental parameter interaction analysis module, configured to perform an interaction analysis between environmental parameters based on implicit information association on the dust concentration time series fluctuation feature and the air humidity time series fluctuation feature to obtain a dust concentration-air humidity time series collaboration feature;

[0038] A time series feature fusion module, configured to fuse the dust concentration time series fluctuation feature, the air humidity time series fluctuation feature, and the dust concentration-air humidity time series collaboration feature to obtain a multi-dimensional time series feature of environmental parameters;

[0039] A deceleration strategy correction module, configured to correct the deceleration strategy based on the multi-dimensional time series feature of environmental parameters to obtain a corrected deceleration.

[0040] Compared with the related art, the anti-collision control system and method for a bucket wheel stacker-reclaimer provided by the present application, after detecting the existence of an obstacle, trigger a corresponding deceleration strategy based on the position of the obstacle, and at the same time, monitor the dust concentration and air humidity in the working environment in real time, and use a time series analysis technology based on deep learning to perform data analysis on the dust concentration and air humidity data to extract the time series fluctuation features of the dust concentration and air humidity, and then capture the collaborative changes between the dust concentration and air humidity by mining the implicit information association between the two, and correct the deceleration strategy based on this, and control the movement of the bucket wheel stacker-reclaimer according to the corrected deceleration. This method can effectively improve the anti-collision performance of the bucket wheel stacker-reclaimer in a complex environment, reduce the collision risk, and ensure the operation safety by comprehensively considering the comprehensive influence of multi-dimensional environmental factors on the laser detection performance. Description of the Drawings

[0041] Figure 1 It is a flowchart of the anti-collision control method for a bucket wheel stacker-reclaimer according to an embodiment of the present application;

[0042] Figure 2Schematic diagram of data flow for the anti-collision control method of a bucket wheel stacker-reclaimer according to an embodiment of the present application;

[0043] Figure 3 Flowchart of sub-step S4 of the anti-collision control method of a bucket wheel stacker-reclaimer according to an embodiment of the present application;

[0044] Figure 4 Flowchart of sub-step S42 of the anti-collision control method of a bucket wheel stacker-reclaimer according to an embodiment of the present application;

[0045] Figure 5 Flowchart of sub-step S43 of the anti-collision control method of a bucket wheel stacker-reclaimer according to an embodiment of the present application;

[0046] Figure 6 Flowchart of sub-step S6 of the anti-collision control method of a bucket wheel stacker-reclaimer according to an embodiment of the present application;

[0047] Figure 7 Block diagram of the anti-collision control system of a bucket wheel stacker-reclaimer according to an embodiment of the present application. Detailed implementation manners

[0048] As shown in the present application and the claims, unless the context clearly indicates otherwise, words such as "a", "an", "one" and / or "the" are not specifically singular and may also include plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0049] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0050] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, according to needs, various steps can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several steps can be removed from these processes.

[0051] Next, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0052] It is worth noting that in this application, all actions to obtain data are carried out in compliance with the relevant data protection laws and policies of the country where the data is located, and with the authorization given by the owner of the corresponding device.

[0053] As mentioned in the background technology above, patent CN115576360A proposes a method for preventing collisions of bucket wheel excavators, which includes: using a laser detection system to scan road condition information within a 270° range. When an obstacle is first detected, the position is recorded, and the bucket wheel excavator speed, acceleration, dust particle content and air humidity data are obtained. The operation state of the bucket wheel excavator is initially adjusted using an adjustment module, and the adjustment is corrected based on environmental data. After an interval of time T1, the position and movement state of the obstacle are detected again, and the operation state of the bucket wheel excavator is adjusted again according to the detection results until the bucket wheel excavator completely passes the obstacle.

[0054] Specifically, the laser detection system is built based on laser ranging technology, which determines the distance and position of the object by emitting a laser beam of a specific wavelength and receiving the reflected signal. In order to adapt to complex working environments, such as high dust concentration and large humidity changes, it is particularly important to choose appropriate laser parameters. Usually, a laser source in the near-infrared band is selected because this band has a strong penetration ability for dust and moisture, which can reduce the impact of external factors on the laser propagation path. In addition, in order to achieve all-round monitoring, the laser launch device is installed on a rotatable platform that can rotate at a large angle in the horizontal direction to achieve the purpose of covering a range of 270°. When the laser beam encounters an obstacle, the exact position of the obstacle is calculated based on the time difference of the reflected light, and this information is transmitted to the central processing unit.

[0055] In actual operation, in order to ensure the accuracy of the data, the laser detection system does not rely solely on a single measurement result, but uses multiple measurements and calculates the average value to improve accuracy. This method effectively reduces errors caused by accidental factors and enhances the system's adaptability to environmental changes. At the same time, considering the changes in laser performance under different environmental conditions, the system is also equipped with an adaptive adjustment mechanism that can automatically adjust the laser's emission power and frequency based on real-time feedback to maintain the best working state. This dynamic adjustment not only improves the reliability of detection, but also provides a more stable foundation for subsequent data analysis.

[0056] The speed, acceleration, dust particle content, and air humidity data of the bucket wheel stacker-reclaimer are important bases for evaluating the current operating environment status and are also key inputs for formulating effective deceleration strategies. Speed sensors and accelerometers are widely used to monitor the motion state of the bucket wheel stacker-reclaimer. Speed sensors are usually designed based on the Hall effect or optical encoders and can record the moving speed of the bucket wheel stacker-reclaimer in real time. The accelerometer, on the other hand, can capture the subtle changes during the acceleration or deceleration process of the machine, which is crucial for determining when to take emergency braking measures. By combining the information provided by both, the control center can comprehensively understand the dynamic behavior of the bucket wheel stacker-reclaimer and thus make more accurate decisions.

[0057] The dust particle content and air humidity are the main indicators reflecting the changes in physical conditions in the operating environment. Dust concentration sensors mostly use the optical scattering method to measure the number density of dust in the air, that is, by measuring the degree of scattering of the incident light by dust particles to estimate the dust concentration. This method has the characteristics of fast response speed and high sensitivity and is very suitable for real-time monitoring. As for humidity sensors, they may use various principles such as capacitive, resistive, or thermal conduction to measure air humidity. Each type of humidity sensor has its unique advantages and applicable scenarios. For example, capacitive humidity sensors are favored in many applications due to their small size and fast response. All these sensors must have good stability and anti-interference ability to operate continuously and reliably in harsh working environments.

[0058] In the related technology, although the influence of dust concentration and air humidity on laser detection is considered, which improves the operating safety of the bucket wheel stacker-reclaimer to a certain extent, its correction strategy is relatively simple and mainly relies on a preset correction coefficient to linearly correct the speed of the bucket wheel stacker-reclaimer. However, in the actual operating environment, the dust concentration and air humidity often show complex dynamic change characteristics, and the two have complex interactive effects on the performance of laser detection. Traditional preset correction coefficients usually cannot accurately reflect this complex interactive effect, resulting in poor correction effects. To address the above technical problems, this application proposes an optimized anti-collision control method for the bucket wheel stacker-reclaimer. After detecting the presence of an obstacle, it triggers a corresponding deceleration strategy based on the position of the obstacle, while simultaneously monitoring the dust concentration and air humidity in the operating environment in real time, and using deep learning-based time series analysis technology to analyze the dust concentration and air humidity data to extract the time series fluctuation characteristics of the dust concentration and air humidity. Furthermore, by mining the implicit information correlation between the two, it captures the co-variation between the dust concentration and air humidity, and based on this, corrects the deceleration strategy to control the movement of the bucket wheel stacker-reclaimer according to the corrected deceleration. By comprehensively considering the comprehensive influence of multi-dimensional environmental factors on the performance of laser detection, this method can effectively improve the anti-collision performance of the bucket wheel stacker-reclaimer in a complex environment, reduce the collision risk, and ensure the operating safety.

[0059] Figure 1 Flowchart of the anti-collision control method for a bucket wheel stacker-reclaimer according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the anti-collision control method for a bucket wheel stacker-reclaimer according to an embodiment of the present application. As Figure 1 and Figure 2 shown, the anti-collision control method for the bucket wheel stacker-reclaimer includes the steps of: S1, if the obstacle is in the front area, trigger an emergency deceleration strategy; if the obstacle is not in the front area, trigger a secondary deceleration strategy and give an alarm. S2, obtain the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value. S3, extract the dust concentration time-series fluctuation feature and the air humidity time-series fluctuation feature from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value. S4, perform an interactive analysis between environmental parameters based on implicit information association on the dust concentration time-series fluctuation feature and the air humidity time-series fluctuation feature to obtain the dust concentration-air humidity time-series collaborative feature. S5, fuse the dust concentration time-series fluctuation feature, the air humidity time-series fluctuation feature, and the dust concentration-air humidity time-series collaborative feature to obtain the multi-dimensional time-series feature of environmental parameters. S6, based on the multi-dimensional time-series feature of environmental parameters, correct the deceleration strategy to obtain the corrected deceleration.

[0060] In the above anti-collision control method for the bucket wheel stacker-reclaimer, in step S1, if the obstacle is in the front area, trigger an emergency deceleration strategy; if the obstacle is not in the front area, trigger a secondary deceleration strategy and give an alarm. Specifically, first, according to the road condition information within the surrounding 270° range obtained by the laser detection system, determine whether the obstacle is in the front area. In an embodiment of the present application, the front area refers to a fan-shaped area of 45° to the left and right of the due front of the driving direction of the bucket wheel stacker-reclaimer. If it is detected that the obstacle is within this front area, it is considered a high-risk situation, and then an emergency deceleration strategy is immediately triggered, and the operating speed of the bucket wheel stacker-reclaimer is quickly reduced through the control system to avoid collision with the obstacle. If the obstacle is not in the front area, although the collision risk is relatively small, attention still needs to be paid. Therefore, a secondary deceleration strategy is triggered, and at the same time, the operator is reminded through an acoustic-optic alarm device to ensure that manual intervention measures can be taken when necessary. In a specific example of the present application, the deceleration of the emergency deceleration strategy and the deceleration of the secondary deceleration strategy are the first deceleration and the second deceleration respectively, and the first deceleration is greater than the second deceleration.

[0061] In the above anti-collision control method for the bucket wheel stacker-reclaimer, in step S2, a time queue of real-time dust concentration values and a time queue of real-time air humidity values are obtained. It should be understood that in this application, considering that in the working environment of the bucket wheel stacker-reclaimer, changes in dust concentration and air humidity will affect the detection system of the bucket wheel stacker-reclaimer, thereby affecting the detection accuracy of obstacles. For example, a higher dust concentration may cause the laser signal of the laser detection system to scatter and attenuate, resulting in inaccurate detected obstacle distances. And a larger air humidity may cause water vapor condensation on the surface of optical components, affecting the normal operation of detection equipment. Therefore, in order to avoid detection errors caused by environmental factors and thus cause inappropriate deceleration decisions, this application further monitors the dust concentration and air humidity in the working environment in real time, obtains a time queue of real-time dust concentration values and a time queue of real-time air humidity values, so as to dynamically adjust the deceleration strategy by considering the timing cumulative effect of environmental factors on laser detection performance, thereby improving the adaptability and accuracy of the anti-collision system.

[0062] Specifically, in order to achieve real-time monitoring of dust concentration and air humidity, a series of high-precision sensors are usually deployed, including dust concentration sensors and humidity sensors. Dust concentration sensors work based on the principle of light scattering, and calculate the dust concentration by measuring the scattering intensity of incident laser by dust particles in the air. Such sensors have the characteristics of fast response and high sensitivity, and are very suitable for dynamically monitoring dust changes in the environment. Specifically, when a laser beam passes through air containing dust particles, the dust particles will cause light scattering. The sensor can calculate the dust concentration in the current environment by detecting the intensity change of the scattered light and combining with a pre-set calibration curve. It should be noted that in actual applications, to ensure the accuracy and reliability of data, dust concentration sensors need to be calibrated regularly to adapt to changes in dust characteristics in different seasons and working conditions.

[0063] For the measurement of air humidity, capacitive or resistive humidity sensors are mostly used. Capacitive humidity sensors determine air humidity by measuring the change in capacitance caused by the change in the dielectric constant of the humidity-sensitive material with humidity. Resistive humidity sensors are designed based on the principle that the resistance value of the humidity-sensitive element changes correspondingly with humidity. No matter which type is used, humidity sensors need to have good linearity and stability to ensure reliable data output under wide-range humidity changes. At the same time, considering factors such as electromagnetic interference that may exist at the operation site, the sensors also need to take effective shielding measures to avoid the influence of external factors on the measurement results.

[0064] During the operation of the bucket wheel stacker-reclaimer, sensors continuously collect relevant parameters in the surrounding environment. However, it is often difficult to directly use the original data to meet the analysis requirements because the data may have problems such as noise interference or inconsistent sampling intervals. Therefore, preprocessing operations are usually performed on the collected data, such as filtering and denoising, and smoothing, to improve the data quality. For example, using a low-pass filter can effectively remove high-frequency noise components, making the signal smoother and more stable. Using the moving average method to smooth the data helps to eliminate the influence of random errors, making the trend more clearly visible.

[0065] Whenever new sensor data is received, the controller will first perform necessary preprocessing operations on it, and then insert it into the corresponding time queue according to the timestamp of the data. It should be noted here that in order to ensure the accuracy and consistency of the time queue, the system must have an accurate time synchronization mechanism. Usually, the Network Time Protocol (NTP) or other high-precision time synchronization schemes are adopted to ensure that the time deviation between all devices is controlled within an acceptable range.

[0066] In the above anti-collision control method for the bucket wheel stacker-reclaimer, in step S3, the time series fluctuation characteristics of the dust concentration and the time series fluctuation characteristics of the air humidity are extracted from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value. It should be understood that since the dust concentration and air humidity in the operating environment are not fixed, but will change continuously with factors such as the passage of time, the progress of operating activities, and external meteorological conditions, and the changes in such environmental parameters may have a cumulative effect, posing a potential threat to the operating safety of the bucket wheel stacker-reclaimer. For example, when in a high dust concentration environment for a long time, dust particles may gradually accumulate on the lens surface of the laser detection system, resulting in a gradual attenuation of the detection signal. And the continuous increase in air humidity may exacerbate the water vapor condensation phenomenon on the surface of optical components, further reducing the sensitivity of the detection system. Therefore, in order to effectively capture the dynamic changes of environmental parameters and more accurately evaluate the impact of environmental factors on laser detection performance, this application adopts a time series analysis technology based on deep learning to perform time series feature extraction on the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value respectively, so as to obtain the time series fluctuation characteristics of the dust concentration and the time series fluctuation characteristics of the air humidity. In a specific example of this application, a time series encoder based on the GRU model is used to perform time series feature extraction on the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value respectively to obtain the time series fluctuation characteristics of the dust concentration and the time series fluctuation characteristics of the air humidity. Those of ordinary skill in the art should know that GRU is a special structure of the recurrent neural network (RNN). By introducing the update gate and reset gate mechanisms to control the flow of information, it effectively solves the problem of gradient disappearance encountered by traditional RNNs during long sequence training. Specifically, when processing the time series data of dust concentration and air humidity, the GRU model can dynamically determine which historical information should be retained or forgotten, and accordingly generate feature representations containing the change rules and fluctuation patterns in the time dimension, that is, the time series fluctuation characteristics of the dust concentration and the time series fluctuation characteristics of the air humidity. This characteristic makes the GRU model particularly suitable for capturing the complex dynamic behavior of environmental parameters evolving over time, providing strong data support for subsequent analysis. In this way, not only can the change trend of the current environmental conditions be identified, but also the possible future development direction can be predicted, so as to make more accurate adjustment decisions.

[0067] In the above anti-collision control method for the bucket-wheel stacker-reclaimer, in step S4, an interactive analysis between environmental parameters based on implicit information association is performed on the temporal fluctuation characteristics of the dust concentration and the temporal fluctuation characteristics of the air humidity to obtain the dust concentration-air humidity temporal coordination characteristics. Specifically, since the dust concentration and the air humidity in the working environment do not exist in isolation, there is a complex internal relationship between the two. For example, changes in air humidity may affect the suspension and settlement characteristics of dust. When the air humidity is high, water vapor will adhere to the surface of dust particles, increasing the mass of the dust particles and thus accelerating their settlement, resulting in a decrease in the dust concentration. Conversely, in a dry environment, dust is more likely to be suspended in the air, and the dust concentration may increase. Therefore, the present application further performs an implicit correlation analysis on the temporal fluctuation characteristics of the dust concentration and the temporal fluctuation characteristics of the air humidity to capture the co-variation between the two, so as to more accurately reflect the comprehensive influence of environmental factors on the laser detection performance. For example, under the combined action of a high dust concentration and a high air humidity, the scattering and attenuation of the laser signal by the laser detection system will be more severe, resulting in a significant decrease in the detection accuracy. Therefore, by analyzing the temporal coordination characteristics between the dust concentration and the air humidity, the present application can more accurately evaluate the degree of influence of the co-variation of multi-dimensional environmental factors on the detection system, thereby providing a more reliable basis for the subsequent correction of the deceleration strategy. Among them, Figure 3 is a flowchart of sub-step S4 of the anti-collision control method for the bucket-wheel stacker-reclaimer according to an embodiment of the present application. As Figure 3 shown, step S4 includes the steps of: S41, performing a principal component analysis on the temporal fluctuation characteristics of the air humidity to obtain a set of principal component components of the temporal fluctuation characteristics of the air humidity. S42, respectively performing an implicit correlation analysis on the temporal fluctuation characteristics of the dust concentration and each of the principal component components of the temporal fluctuation characteristics of the air humidity in the set to construct a {deep implicit feature of the temporal fluctuation of the dust concentration, anchored deep implicit feature of the principal component of the temporal fluctuation of the air humidity} feature pair. S43, performing a fine-grained interactive analysis on the {deep implicit feature of the temporal fluctuation of the dust concentration, anchored deep implicit feature of the principal component of the temporal fluctuation of the air humidity} feature pair to obtain the dust concentration-air humidity temporal coordination characteristics.

[0068] Specifically, step S41 is expressed by the formula:

[0069]

[0070] where C 2 is the covariance matrix of the temporal fluctuation characteristics of the air humidity, and U 2 represents a matrix composed of a set of principal component components of the temporal fluctuation characteristics of the air humidity obtained by performing eigenvalue decomposition on C 2 , and (·) Tdenotes transpose, Λ 2 denotes performing on C 2 eigenvalue decomposition to obtain a diagonal matrix composed of a set of principal component eigenvalues of the temporal fluctuation of air humidity, λ 21 and λ 2m respectively denote the 1st and the m-th eigenvalues in the set of principal component eigenvalues of the temporal fluctuation of air humidity, v 21 、v 22 and v 2m respectively denote the 1st, the 2nd, and the m-th principal component components of the temporal fluctuation characteristics of air humidity in U 2 , where m is the number of the principal component components of the temporal fluctuation characteristics of air humidity, PCA(·) denotes the principal component analysis network, and V 2 denotes the temporal fluctuation characteristics of air humidity.

[0071] That is, considering that under normal circumstances, the temporal variation of air humidity is relatively stable compared to the dust concentration, and its temporal fluctuation characteristics may contain more redundant information. Therefore, in this application, the principal component analysis is first performed on the temporal fluctuation characteristics of air humidity to construct a new low-dimensional feature space, and the temporal fluctuation characteristics of air humidity are transformed into a more representative set of principal component components of the temporal fluctuation characteristics of air humidity, so as to facilitate the extraction of the main change trend and key information in the temporal fluctuation characteristics of air humidity, eliminate redundancy and noise, and reduce the complexity and computational amount of subsequent interactive analysis.

[0072] Figure 4 is a flowchart of sub-step S42 of the anti-collision control method for a bucket wheel stacker-reclaimer according to an embodiment of this application. As Figure 4 shown, the step S42 includes steps: S421, performing implicit feature extraction based on point convolution coding on the temporal fluctuation characteristics of the dust concentration to obtain deep implicit features of the temporal fluctuation of the dust concentration. S422, performing implicit feature extraction based on point convolution coding on each principal component component of the temporal fluctuation characteristics of air humidity in the set of principal component components of the temporal fluctuation characteristics of air humidity to obtain a set of deep implicit features of the principal components of the temporal fluctuation of air humidity. S423, inputting the deep implicit features of the temporal fluctuation of the dust concentration and the set of deep implicit features of the principal components of the temporal fluctuation of air humidity into an implicit key clue anchoring network to obtain the {deep implicit features of the temporal fluctuation of the dust concentration, anchored deep implicit features of the principal components of the temporal fluctuation of air humidity} feature pair.

[0073] More specifically, the step S421 is expressed by the formula:

[0074] H 1 = sigmoid[Conv 1×1 (V 1 )]

[0075] Among them, V 1 represents the time - series fluctuation characteristics of the dust concentration, Conv 1×1 (·) represents the point - convolution operation, sigmoid(·) represents the sigmoid activation function, and H 1 represents the deep implicit characteristics of the time - series fluctuation of the dust concentration.

[0076] More specifically, the step S422 is expressed by the formula:

[0077]

[0078] Among them, v 2i represents the i - th main - component component of the time - series fluctuation characteristics of the air humidity in the set of the main - component components of the time - series fluctuation characteristics of the air humidity, and H 2 represents the set of deep implicit characteristics of the main - component of the time - series fluctuation of the air humidity, and h 21 , h 22 , h 2i and h 2m respectively represent the 1st, 2nd, i - th, and m - th deep implicit characteristics of the main - component of the time - series fluctuation of the air humidity in the set of the deep implicit characteristics of the main - component of the time - series fluctuation of the air humidity.

[0079] That is, the present application further extracts the deep implicit characteristics of the time - series fluctuation characteristics of the dust concentration and each main - component component of the time - series fluctuation characteristics of the air humidity through point - convolution encoding processing. In particular, point - convolution does not focus on the locality of the original space of the features, but directly performs information fusion within the global scope of the features through cross - dimensional weight mapping, and further enhances the feature expression ability through a non - linear activation function, so as to be able to efficiently extract deep - level and strongly representative implicit features, obtaining a set of deep implicit characteristics of the time - series fluctuation of the dust concentration and deep implicit characteristics of the main - component of the time - series fluctuation of the air humidity.

[0080] More specifically, in a specific example of the present application, the step S423 includes: calculating the feature correlation factors between the deep implicit characteristics of the time - series fluctuation of the dust concentration and each deep implicit characteristic of the main - component of the time - series fluctuation of the air humidity in the set of the deep implicit characteristics of the main - component of the time - series fluctuation of the air humidity, and selecting the deep implicit characteristic of the main - component of the time - series fluctuation of the air humidity corresponding to the largest feature correlation factor as the anchored deep implicit characteristic of the main - component of the time - series fluctuation of the air humidity, so as to construct the {deep implicit characteristics of the time - series fluctuation of the dust concentration, anchored deep implicit characteristic of the main - component of the time - series fluctuation of the air humidity} feature pair, which is expressed by the formula:

[0081]

[0082] F bestpair= {H 1 ; h 2k}

[0083] where <·> represents the calculation of the inner product, ε represents the smoothing coefficient, ‖·‖ represents the calculation of the norm, arg max(·) represents the index corresponding to the maximum value, k represents the index of the deep hidden feature of the main component of the temporal fluctuation of air humidity corresponding to the largest characteristic correlation factor, and h 2k represents the deep hidden feature of the main component of the temporal fluctuation of the anchored air humidity, and F bestpair represents the feature pair {the deep hidden feature of the temporal fluctuation of dust concentration, the deep hidden feature of the main component of the temporal fluctuation of the anchored air humidity}.

[0084] That is, in order to filter out the noise and redundant components in the set of deep hidden features of the main component of the temporal fluctuation of air humidity, the present application further uses an implicit key clue anchoring network to perform implicit correlation analysis on the set of deep hidden features of the temporal fluctuation of dust concentration and the deep hidden features of the main component of the temporal fluctuation of air humidity, so as to capture the implicit key clues between the temporal fluctuations of dust concentration and air humidity, in order to achieve semantic alignment of environmental parameter features from different sources. In a specific implementation, by calculating the correlation between the deep hidden feature of the temporal fluctuation of dust concentration and each deep hidden feature of the main component of the temporal fluctuation of air humidity, the most relevant deep hidden feature of the main component of the temporal fluctuation of air humidity is extracted as the key clue, and it is used as the anchor point to construct the feature pair {the deep hidden feature of the temporal fluctuation of dust concentration, the deep hidden feature of the main component of the temporal fluctuation of the anchored air humidity}, so as to endow the feature interaction with a clear structural prior through the alignment operation and eliminate the ambiguity between different feature sources.

[0085] Preferably,[[]] Figure 5 is a flowchart of sub-step S43 of the anti-collision control method for a bucket wheel stacker-reclaimer according to an embodiment of the present application. As Figure 5 shown, the step S43 includes the steps: S431, performing feature power-law distribution alignment constraint on the deep hidden feature of the temporal fluctuation of dust concentration and the deep hidden feature of the main component of the temporal fluctuation of the anchored air humidity to obtain the feature pair {the optimized deep hidden feature of the temporal fluctuation of dust concentration, the optimized deep hidden feature of the main component of the temporal fluctuation of the anchored air humidity}. S432, inputting the feature pair {the optimized deep hidden feature of the temporal fluctuation of dust concentration, the optimized deep hidden feature of the main component of the temporal fluctuation of the anchored air humidity} into the feature fine-grained interaction network to obtain the dust concentration-air humidity temporal coordination feature.

[0086] More specifically, the step S431 is expressed by the formula:[[]]

[0087]

[0088] Among them, e (·) represents the exponential function with the natural constant as the base, H 1i and h 2ki are respectively the eigenvalues at the i-th position in the vectors H 1 and j 2k ; H 1i ' represents the optimized deep implicit features of the temporal fluctuation of the dust concentration, and h 2ki ' represents the optimized deep implicit features of the principal component of the temporal fluctuation of the anchored air humidity.

[0089] Specifically, in view of the alignment ambiguity caused by source uncertainty between the high-dimensional features of dust concentration and air humidity from different sources in this application, a weakening and blurring mechanism is introduced to perform weak blurring power-law expansion. That is, the 3 / 8 exponent is used as the precursor prior expansion to perform power-law prior responsive blurring convergence on the feature parameter boundary alignment condition, and the 1 / 4 exponent is used as the main body alignment expansion to perform strict constraint on the relaxation of the alignment attenuation of the feature power-law distribution. Thus, in the case where the alignment boundary condition constraint within the correlation effective range is poorly defined, the value correlation mechanism is used to avoid the prior ambiguity of the system behavior under a single mechanism, so as to correct the semantic distribution consistency blurring within the alignment interval and enhance the intuitiveness of mining the implicit fine-grained interaction correlation features between the anchored feature pairs.

[0090] More specifically, the step S432 is expressed by the formula:

[0091]

[0092] Among them, softmax(·) represents the softmax function, α and β respectively represent different weight parameters, S represents the feature scale scaling parameter, and V f represents the dust concentration-air humidity temporal collaborative feature.

[0093] That is, the anchored feature pairs are input into the feature fine-grained interaction network, and more complex non-linear feature correlations between the dust concentration feature and the air humidity feature are mined on the basis of the existing semantic alignment. In specific implementation, the feature fine-grained interaction network is based on the self-attention mechanism. By calculating the attention weights between the feature pairs, the correlation strength between different features is dynamically adjusted to capture a more refined feature interaction pattern between the dust concentration and the air humidity, and improve the accuracy and efficiency of feature interaction analysis. In addition, the feature fine-grained interaction network further enhances the integrity of the feature representation by introducing the original temporal fluctuation features of the dust concentration and the principal component components of the temporal fluctuation features of the anchored air humidity, and obtains the dust concentration-air humidity temporal collaborative feature, which helps to more comprehensively understand the comprehensive effect of the co-variation between the dust concentration and the air humidity on the laser detection performance.

[0094] In the above anti-collision control method for the bucket wheel stacker-reclaimer, in step S5, the time-series fluctuation characteristics of the dust concentration, the time-series fluctuation characteristics of the air humidity, and the time-series collaborative characteristics of the dust concentration-air humidity are fused to obtain the multi-dimensional time-series characteristics of the environmental parameters. It should be understood that the dust concentration and the air humidity are two key environmental factors affecting the laser detection performance. Their respective time-series fluctuation characteristics and the collaborative characteristics between them together constitute a comprehensive description of the impact on the laser detection performance. Therefore, in this application, by further fusing the time-series fluctuation characteristics of the dust concentration, the time-series fluctuation characteristics of the air humidity, and the time-series collaborative characteristics of the dust concentration-air humidity, the independent change information of the dust concentration and the air humidity and the collaborative change between them are comprehensively considered, and the actual situation in the current working environment is more comprehensively revealed, providing richer information support for the subsequent correction of the deceleration strategy. In specific implementation, methods such as feature splicing and weighted summation can be used to achieve the fusion of features, so as to obtain the multi-dimensional time-series characteristics of the environmental parameters containing more effective information.

[0095] In the above anti-collision control method for the bucket wheel stacker-reclaimer, in step S6, based on the multi-dimensional time-series characteristics of the environmental parameters, the deceleration strategy is corrected to obtain the corrected deceleration. Among them, Figure 6 FIG. is a flowchart of sub-step S6 of the anti-collision control method for the bucket wheel stacker-reclaimer according to an embodiment of the present application. As Figure 6 shown, step S6 includes steps: S61, inputting the multi-dimensional time-series characteristics of the environmental parameters into a correction coefficient estimation module based on a decoder to obtain an environmental impact correction coefficient. S62, multiplying the environmental impact correction coefficient by the first deceleration or the second deceleration to obtain the corrected deceleration.

[0096] Specifically, in step S61, the multi-dimensional time-series characteristics of the environmental parameters are input into a correction coefficient estimation module based on a decoder to obtain an environmental impact correction coefficient. Those of ordinary skill in the art should know that the decoder model is a neural network structure used to map the input encoded features to the output space and has powerful learning ability. In this application, the decoder model is used to receive the multi-dimensional time-series characteristics of the environmental parameters as input, and by learning the complex correlation relationship between the environmental characteristics and the laser detection performance, map the multi-dimensional time-series characteristics of the environmental parameters into an environmental impact correction coefficient, so as to quantitatively represent the specific impact degree of the dust concentration and the air humidity in the current working environment on the laser detection performance.

[0097] Specifically, in step S62, multiply the environmental impact correction coefficient by the first deceleration or the second deceleration to obtain the corrected deceleration. That is, multiply the environmental impact correction coefficient by the initial deceleration (i.e., the first deceleration or the second deceleration) obtained based on the obstacle position information to compensate for the potential impact of environmental factors on the laser detection performance, dynamically adjust the deceleration strategy, and thus obtain a corrected deceleration that better conforms to the current environmental conditions. For example, when the environmental humidity is high and the dust concentration is large, the laser detection performance may be greatly affected, resulting in a shortened detection distance or a decreased detection accuracy. At this time, the environmental impact correction coefficient estimated by the decoder model will increase accordingly, and then the corrected deceleration will also increase accordingly to ensure that the bucket wheel stacker-reclaimer travels at a safer speed and avoid potential collision risks caused by detection errors due to environmental factors.

[0098] In the above anti-collision control method for the bucket wheel stacker-reclaimer, the process of detecting the obstacle position and motion state again after the interval time T1 and adjusting the operating state according to the detection results is to ensure the safe passage of the equipment through the obstacle. Specifically, after the obstacle is first detected and its position is recorded, the system will wait for a preset time interval T1. The selection of this time is crucial. It should be long enough to allow the bucket wheel stacker-reclaimer to have time to respond to the initial detection result, and as short as possible to quickly adapt to environmental changes. During this time, the bucket wheel stacker-reclaimer will be adjusted according to the corrected rear deceleration strategy, and the central control system continuously monitors parameters such as speed, acceleration, dust concentration, and air humidity. These real-time data provide the necessary input conditions for subsequent adjustments.

[0099] After reaching the T1 moment, the laser detection system restarts and begins a new round of scanning work. The laser beam emission frequency and power may be adjusted according to the results of the previous environmental data analysis at this stage to optimize the detection effect. For example, in a high-dust or high-humidity environment, a higher-frequency pulse signal may be used to improve the resolution. In a relatively clear environment, the frequency can be appropriately reduced to save energy. In addition, to further improve the detection accuracy, the system may also introduce a multi-angle scanning mode to perform multiple detections on the same area from different directions, thereby forming a more three-dimensional image.

[0100] As the laser beam continuously emits outward and receives reflected signals, a large amount of information about the position and shape of obstacles is quickly collected. At this time, after receiving this raw data, the central processing unit immediately starts preliminary processing. The main tasks at this stage include removing noise interference, correcting measurement deviations caused by environmental factors, etc. Specifically, by comparing the previous scan results and using algorithms such as differential filters to eliminate outliers, the authenticity and reliability of the data used are ensured. Meanwhile, the sensor continues to monitor the speed and acceleration changes of the bucket wheel stacker-reclaimer. These dynamic parameters not only reflect the current state of the equipment but also provide an important basis for evaluating whether obstacle avoidance is successful. For example, if it is found that the speed reduction exceeds the expected value, it indicates that there may be unrecognized risk points. On the contrary, if the speed returns to the normal level, it means that the obstacle has been effectively avoided.

[0101] In summary, the anti-collision control method for a bucket wheel stacker-reclaimer based on the embodiments of the present application is elucidated. After detecting the existence of an obstacle, it triggers a corresponding deceleration strategy based on the position of the obstacle, while simultaneously monitoring the dust concentration and air humidity in the working environment in real time, and using time series analysis technology based on deep learning to perform data analysis on the dust concentration and air humidity data to extract the time series fluctuation characteristics of the dust concentration and air humidity. Furthermore, by mining the implicit information correlation between the two, the co-variation between the dust concentration and air humidity is captured, and based on this, the deceleration strategy is corrected to control the movement of the bucket wheel stacker-reclaimer according to the corrected deceleration. By comprehensively considering the comprehensive influence of multi-dimensional environmental factors on the laser detection performance, this method can effectively improve the anti-collision performance of the bucket wheel stacker-reclaimer in a complex environment, reduce the collision risk, and ensure the operation safety.

[0102] Furthermore, the present application also provides an anti-collision control system for a bucket wheel stacker-reclaimer.

[0103] Figure 7 It is a block diagram of the anti-collision control system for a bucket wheel stacker-reclaimer according to the embodiments of the present application. As Figure 7As shown, the anti-collision control system 100 of a bucket-wheel stacker-reclaimer according to an embodiment of the present application includes: an obstacle detection module 110, which is used to scan the road conditions within a range of 270° around by using a laser detection system. If an obstacle is located in the front area, an emergency deceleration strategy is triggered. If the obstacle is not in the front area, a secondary deceleration strategy is triggered and an alarm is given. A real-time data acquisition module 120, which is used to acquire the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value. A timing feature extraction module 130, which is used to extract the dust concentration timing fluctuation feature and the air humidity timing fluctuation feature from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value. An environmental parameter interaction analysis module 140, which is used to perform an interaction analysis between environmental parameters based on implicit information association on the dust concentration timing fluctuation feature and the air humidity timing fluctuation feature to obtain a dust concentration-air humidity timing collaboration feature. A timing feature fusion module 150, which is used to fuse the dust concentration timing fluctuation feature, the air humidity timing fluctuation feature, and the dust concentration-air humidity timing collaboration feature to obtain a multi-dimensional timing feature of environmental parameters. A deceleration strategy correction module 160, which is used to correct the deceleration strategy based on the multi-dimensional timing feature of environmental parameters to obtain a corrected deceleration.

[0104] Here, those skilled in the art can understand that the specific operations of each module in the above anti-collision control system of the bucket-wheel stacker-reclaimer have been introduced in detail in the description of the Figures 1 to 6 anti-collision control method of the bucket-wheel stacker-reclaimer above, and therefore, the repeated description thereof will be omitted.

[0105] The basic principles of the present invention have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, advantages, effects, etc. mentioned in the present invention are only examples and not limitations. It cannot be considered that these advantages, advantages, effects, etc. are essential for each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and easy understanding, rather than limitations. The above details do not limit the present invention to necessarily adopt the above specific details to implement.

[0106] In the above embodiments, the descriptions of the respective embodiments have their own focuses. For parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0107] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

[0108] In addition, it is obvious that the word "including" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0109] Finally, it should be noted that the above description has been given for purposes of illustration and description. In addition, the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for preventing collision of a bucket wheel stacker and reclaimer, comprising: The laser detection system is used to scan the road condition information within a 270° range around; the position is recorded when the obstacle is detected for the first time, and the bucket wheel machine speed, acceleration, dust particle content and air humidity data are obtained; the adjustment module is used to adjust the bucket wheel machine running state for the first time, and the adjustment is corrected based on the environmental data; the obstacle position and movement state are detected again after an interval of T1, and the bucket wheel machine running state is adjusted again according to the detection result until the bucket wheel machine completely passes the obstacle, characterized in that the adjustment module is used to adjust the bucket wheel machine running state for the first time, and the adjustment is corrected based on the environmental data, including: If the obstacle is in the front area, the emergency deceleration strategy is triggered; if the obstacle is not in the front area, the secondary deceleration strategy is triggered and an alarm is sounded; Obtain the time queue of real-time dust concentration value and the time queue of real-time air humidity value; Extracting dust concentration time series fluctuation characteristics and air humidity time series fluctuation characteristics from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value; Performing an interactive analysis between environmental parameters based on implicit information association on the dust concentration time series fluctuation characteristics and the air humidity time series fluctuation characteristics to obtain dust concentration-air humidity time series synergistic characteristics; The dust concentration time series fluctuation characteristics, the air humidity time series fluctuation characteristics and the dust concentration-air humidity time series synergy characteristics are integrated to obtain a multi-dimensional time series characteristic of an environmental parameter; Based on the multi-dimensional time series characteristics of the environmental parameters, the deceleration strategy is modified to obtain a modified deceleration.

2. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 1, characterized in that: The deceleration of the emergency deceleration strategy and the deceleration of the secondary deceleration strategy are respectively a first deceleration and a second deceleration, and the first deceleration is greater than the second deceleration.

3. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 2, characterized in that: Extracting dust concentration time series fluctuation characteristics and air humidity time series fluctuation characteristics from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value includes: A time series encoder based on a GRU model is used to extract time series features from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value respectively to obtain the time series fluctuation features of the dust concentration and the time series fluctuation features of the air humidity.

4. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 3, characterized in that: The dust concentration time series fluctuation characteristics and the air humidity time series fluctuation characteristics are interactively analyzed based on implicit information association between environmental parameters to obtain dust concentration-air humidity time series synergistic characteristics, including: Performing characteristic principal component analysis on the air humidity time series fluctuation characteristics to obtain a set of principal component components of the air humidity time series fluctuation characteristics; Performing implicit correlation analysis on each of the main component components of the air humidity time series fluctuation characteristics in the set of the dust concentration time series fluctuation characteristics and the air humidity time series fluctuation characteristics to construct a feature pair of {deep implicit feature of dust concentration time series fluctuation, anchored deep implicit feature of air humidity time series fluctuation main component}; A fine-grained interactive analysis is performed on the feature pair of {deep implicit features of dust concentration time series fluctuations, deep implicit features of anchored air humidity time series fluctuation principal components} to obtain the dust concentration-air humidity time series synergistic features.

5. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 4, characterized in that: The dust concentration time series fluctuation feature and each of the air humidity time series fluctuation feature principal component components in the set of the air humidity time series fluctuation feature principal component components are subjected to implicit correlation analysis to construct a feature pair of {dust concentration time series fluctuation deep implicit feature, anchored air humidity time series fluctuation principal component deep implicit feature}, including: Performing implicit feature extraction based on point convolution coding on the dust concentration time series fluctuation characteristics to obtain deep implicit features of the dust concentration time series fluctuation; Performing implicit feature extraction based on point convolution coding on each of the main component components of the air humidity time series fluctuation characteristics in the set of main component components of the air humidity time series fluctuation characteristics to obtain a set of deep implicit features of the main component of the air humidity time series fluctuation characteristics; The set of the deep implicit features of the dust concentration time series fluctuation and the deep implicit features of the main components of the air humidity time series fluctuation are input into the implicit key clue anchoring network to obtain the feature pair of {deep implicit features of the dust concentration time series fluctuation, anchored deep implicit features of the main components of the air humidity time series fluctuation}.

6. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 5, characterized in that: Inputting the set of the deep implicit features of the dust concentration time series fluctuation and the deep implicit features of the main component of the air humidity time series fluctuation into the implicit key clue anchoring network to obtain the {deep implicit features of the dust concentration time series fluctuation, anchored deep implicit features of the main component of the air humidity time series fluctuation} feature pair, including: The feature correlation factor between the deep implicit feature of the time series fluctuation of dust concentration and each deep implicit feature of the time series fluctuation of air humidity main component in the set of deep implicit features of the time series fluctuation of air humidity main component is calculated, and the deep implicit feature of the time series fluctuation of air humidity main component corresponding to the largest feature correlation factor is selected as the anchored deep implicit feature of the time series fluctuation of air humidity main component, so as to construct the feature pair of {deep implicit feature of time series fluctuation of dust concentration, anchored deep implicit feature of time series fluctuation of air humidity main component}.

7. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 6, characterized in that: The feature pair {deep implicit features of dust concentration time series fluctuation, deep implicit features of anchored air humidity time series fluctuation principal component} is subjected to fine-grained interactive analysis to obtain the dust concentration-air humidity time series collaborative features, including: Performing feature power-law distribution alignment constraints on the deep implicit features of the dust concentration time series fluctuation and the anchored deep implicit features of the main components of the air humidity time series fluctuation to obtain a feature pair of {optimized deep implicit features of the dust concentration time series fluctuation, optimized deep implicit features of the main components of the anchored air humidity time series fluctuation}; The feature pair of {optimized deep implicit features of dust concentration time series fluctuation, optimized deep implicit features of anchored air humidity time series fluctuation principal components} is input into the feature fine-grained interaction network to obtain the dust concentration-air humidity time series collaborative features.

8. The anti-collision control method for a bucket wheel stacker and reclaimer according to claim 7, characterized in that: Based on the multi-dimensional time series characteristics of the environmental parameters, the deceleration strategy is modified to obtain a modified deceleration, including: Inputting the multi-dimensional time series characteristics of the environmental parameters into a correction coefficient estimation module based on a decoder to obtain an environmental impact correction coefficient; The environmental influence correction coefficient is multiplied by the first deceleration or the second deceleration to obtain the corrected deceleration.

9. A bucket wheel stacker and reclaimer anti-collision control system, characterized in that: include: The obstacle detection module is used to use the laser detection system to scan the road condition information within a 270° range. If the obstacle is located in the front area, the emergency deceleration strategy is triggered. If the obstacle is not in the front area, the secondary deceleration strategy is triggered and an alarm is sounded; A real-time data acquisition module, used to acquire a time queue of real-time dust concentration values ​​and a time queue of real-time air humidity values; A time series feature extraction module, used to extract dust concentration time series fluctuation features and air humidity time series fluctuation features from the time queue of the real-time dust concentration value and the time queue of the real-time air humidity value; An environmental parameter interaction analysis module, used for performing an interaction analysis between environmental parameters based on implicit information association on the dust concentration time series fluctuation characteristics and the air humidity time series fluctuation characteristics to obtain a dust concentration-air humidity time series synergistic characteristic; A time series feature fusion module, used to fuse the dust concentration time series fluctuation feature, the air humidity time series fluctuation feature and the dust concentration-air humidity time series synergy feature to obtain a multi-dimensional time series feature of an environmental parameter; The deceleration strategy correction module is used to correct the deceleration strategy based on the multi-dimensional time series characteristics of the environmental parameters to obtain a corrected deceleration.

Images