Anti-coal-blocking purging flow-aiding intelligent system of coal conveying system

Abstract

The invention discloses an anti-blocking coal purging flow-aiding intelligent system for a coal conveying system, which relates to the field of intelligent operation and maintenance of the coal conveying system and comprises the following components: a sensing and data processing module, a digital twinning construction and simulation module, a remote control and debugging module, an intelligent fault diagnosis module and a man-machine interaction module. According to the invention, through the digital twinborn construction and simulation module, a digital twinborn model for accurately simulating the physical structure, the equipment layout and the operation logic of the coal conveying system is constructed based on a three-dimensional modeling technology, and virtual simulation is carried out on possible coal blockage scenes in combination with historical data, real-time monitoring data and a preset coal blockage scene simulation algorithm; according to the method, the physical characteristics of the coal flow and the coal blockage probability are comprehensively considered, the coal blockage scene is simulated more truly, and the accuracy and efficiency of coal blockage early warning are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent operation and maintenance of coal conveying systems, and specifically to an intelligent system for preventing coal blockage, purging, and flow assistance in coal conveying systems. Background Art

[0002] During the operation of coal conveying systems, coal blockage is a common and troublesome problem. Coal blockage not only reduces the coal conveying efficiency and affects the normal operation of the entire power plant or coal processing facility, but may also cause equipment failures and even safety accidents in severe cases.

[0003] There are many deficiencies in traditional coal conveying system monitoring and anti-blockage technologies. Firstly, the manual inspection method is limited by manpower and time, making it difficult to comprehensively and real-time monitor the coal conveying system, and potential coal blockage hazards are easily overlooked. Secondly, although regular maintenance can prevent problems such as coal blockage to a certain extent, this method often lacks pertinence and is difficult to flexibly adjust according to the actual operating status of the coal conveying system, resulting in high maintenance costs and limited effects. In addition, traditional technologies mainly rely on empirical judgment and simple sensor monitoring for coal blockage early warning, lacking scientific prediction models and data analysis means, and it is difficult to ensure the accuracy and timeliness of early warning.

[0004] In view of the deficiencies of traditional coal conveying system monitoring and anti-blockage technologies, it is particularly important to develop an intelligent system for preventing coal blockage, purging, and flow assistance in coal conveying systems. Summary of the Invention

[0005] The purpose of the present invention is to make up for the deficiencies of the existing technology by providing an intelligent system for preventing coal blockage, purging, and flow assistance in coal conveying systems. It can achieve comprehensive and real-time monitoring and scientific analysis of the operating status of the coal conveying system by integrating multiple advanced components such as a perception and data processing module, a digital twin construction and simulation module, a remote control and debugging module, and an intelligent fault diagnosis module. Especially through the digital twin construction and simulation module, the system can build a digital twin model that accurately simulates the physical structure, equipment layout, and operating logic of the coal conveying system based on three-dimensional modeling technology, and combine historical data, real-time monitoring data, and preset coal blockage scenario simulation algorithms to virtually simulate possible coal blockage scenarios, thereby predicting the possibility, location, and influence range of coal blockage occurrence, and issuing early warnings in a timely manner.

[0006] To solve the above technical problems, the present invention provides the following technical solution: An intelligent system for preventing coal blockage, purging, and flow assistance in coal conveying systems, which includes the following components: a perception and data processing module, a digital twin construction and simulation module, a remote control and debugging module, an intelligent fault diagnosis module, and a human-machine interaction module;

[0007] The perception and data processing module: Deploy various sensors such as pressure, flow, temperature, and vibration at key parts of the coal conveying pipeline, coal dropping pipe, and coal feeder in the coal conveying system to collect equipment operation parameters in real time. The collected data is transmitted to this module through the communication network, and after cleaning, filtering, and analysis preprocessing, the effective information is extracted and stored in the database. At the same time, based on preset rules and algorithms, the equipment operation status is initially evaluated to determine whether there are abnormalities, providing a data basis for subsequent operations;

[0008] The digital twin construction and simulation module: Build a digital twin model based on 3D modeling technology to accurately simulate the physical structure, equipment layout, and operation logic of the coal conveying system, and interact with the perception and data processing module to map the equipment operation data to the model in real time. Using this model, combined with historical data, real-time monitoring data, and preset coal blockage scenario simulation algorithms, virtual simulation of possible coal blockage scenarios is carried out to predict the possibility, location, and influence range of coal blockage, and warnings are issued in a timely manner;

[0009] The remote control and debugging module: Maintenance personnel log in to the system through remote terminal devices such as computers and mobile phones to view the physical equipment status and virtual simulation coal blockage scenarios mapped in real time by the digital twin model. Based on the actual situation, remotely adjust the pressure, time, and interval parameters of the purging system, and this module transmits the adjusted parameters to the purging control system to achieve remote control;

[0010] The intelligent fault diagnosis module: Relying on the digital twin model, real-time monitoring data, and virtual simulation results, use artificial intelligence algorithms and fault diagnosis models to diagnose faults in the coal conveying system. Once abnormal equipment operation or coal blockage is detected, quickly locate the fault location, analyze the cause, and generate a detailed fault diagnosis report, which is displayed through the system interface or sent to relevant personnel via text message or email;

[0011] The human-computer interaction module: Provide an intuitive and friendly user interface. Maintenance personnel can use this to view the real-time operation status of the coal conveying system, digital twin model, virtual simulation results, and fault diagnosis report information, and can also input operation instructions to achieve various controls of the system.

[0012] Furthermore, in the perception and data processing module, the evaluation of the equipment operation status adopts a multi-parameter fusion evaluation algorithm based on dynamic weights. The formula of this algorithm is:

[0013] S = ∑w i (t) × x i

[0014] where S is the comprehensive evaluation value of the equipment operation status, n is the number of types of collected parameters, x i is the normalized real-time value of the i-th parameter, w i(t) is the dynamic weight of the i-th parameter at time t, and the dynamic weight w i (t) is determined as follows: First, based on historical data and expert experience, determine the basic weight w of each parameter's influence on the equipment state under different working conditions i0 , then, introduce the time decay factor α and the parameter change rate influence factor β i , where Δt is the time interval, is the change rate of the i-th parameter within Δt time, and β i is obtained by training the historical fault data through a machine learning algorithm according to the influence degree of parameter change on the equipment state. α is optimized and determined through multiple simulation experiments with the goal of the highest evaluation accuracy. This algorithm can more accurately reflect the actual operating state of the equipment by dynamically adjusting the parameter weights, effectively improving the accuracy of anomaly detection.

[0015] Furthermore, in the digital twin construction and simulation module, the coal blockage scenario simulation algorithm adopts a hybrid model based on fluid dynamics and probability prediction. First, based on the Navier-Stokes equation, establish the basic fluid dynamics model of the coal flow in the coal conveying pipeline, which describes the distribution of physical quantities such as the velocity and pressure of the coal flow. The formula is: where ρ is the coal flow density, is the coal flow velocity vector, p is the pressure, μ is the dynamic viscosity, is the external force. Then, introduce the probability correction factor P i , and establish the coal blockage probability prediction model where P is the total probability of coal blockage occurring in a certain area, m is the number of factors affecting coal blockage, and P j is the probability of coal blockage caused by the j-th factor, and P j is obtained by training the historical coal blockage data through the Bayesian network algorithm in combination with the current equipment operation parameters. This hybrid model comprehensively considers the physical characteristics of the coal flow and the coal blockage probability, can more realistically simulate the coal blockage scenario, and improve the reliability of early warning.

[0016] Furthermore, in the remote control and debugging module, there is an intelligent optimization sub-module for purging parameters. This sub-module uses a hybrid optimization algorithm based on particle swarm optimization and fuzzy logic to determine the purging parameters. First, establish the purging effect evaluation function E = ω1×T + ω2×P + ω3×S, where T is the purging time, P is the purging pressure, S is the degree of coal blockage relief, and ω1, ω2, ω3 are weight coefficients. The weight coefficients are determined by pairwise comparison according to the importance of factors such as operation and maintenance cost, equipment loss, and anti-blockage effect through the analytic hierarchy process. The particle swarm optimization algorithm is used to search for the optimal solution in the parameter space. The particle position update formula is: x id (t + 1) = xid (t) + v id (t + 1), v id (t + 1) = w × v id (t) + c1r1(t)[p id -x id (t)] + c2r2(t)[p gd -x id (t)], where x id is the position of particle i in the d-th dimension, v id is the velocity of particle i in the d-th dimension, w is the inertia weight, c1, c2 are learning factors, r1, r2 are random numbers between [0, 1], p id is the personal best position of particle i, p gd is the global best position. Fuzzy logic is used to dynamically adjust the parameters of the particle swarm optimization algorithm according to the real-time coal jamming situation and equipment status, making the adjustment of the purging parameters more intelligent and efficient.

[0017] Furthermore, in the intelligent fault diagnosis module, a fault classification and location algorithm based on deep learning is adopted to construct a convolutional neural network model. The input layer receives multi-source data including the preprocessed data from the perception and data processing module and the simulation data from the digital twin construction and simulation module. The convolutional layer extracts data features through convolutional kernels, and the formula is:

[0018]

[0019] where is the j-th feature map of the l-th layer, M j is the set of input feature maps, is the i-th feature map of the (l - 1)-th layer, is the convolutional kernel connecting the i-th input feature map and the j-th output feature map of the l-th layer, is the bias, f is the activation function. The pooling layer performs dimensionality reduction on the feature maps, and the fully connected layer integrates the extracted features. The output layer classifies the fault types through the softmax function, and the formula is:

[0020]

[0021] where σ(z) j is the probability of the j-th type of fault, z j is the j-th element of the input vector z, K is the total number of fault types. The model training data comes from the actual coal conveying system fault data and simulated fault data, and the network parameters are adjusted through the backpropagation algorithm to enable the model to accurately identify the fault types and locate the fault positions.

[0022] Furthermore, the human-computer interaction module adopts augmented reality technology. By wearing AR devices, operation and maintenance personnel can intuitively view the superimposed display of the digital twin model of the coal conveying system in the real scene. The operating parameters and fault warning information content of the equipment are marked in real time in the model. When performing remote debugging operations, the AR interface can adjust the parameters of the purging system in the digital twin model through gesture recognition and voice command interaction. At the same time, the AR technology also supports generating three-dimensional visual fault diagnosis reports, presenting the fault location and cause information in the form of a three-dimensional model and animation demonstration, which is convenient for operation and maintenance personnel to understand the fault situation more clearly and accurately, and improves the operation and maintenance efficiency and operation convenience.

[0023] Furthermore, in the perception and data processing module, the data communication adopts a hybrid communication method combining 5G and industrial Ethernet. At the equipment site, the data collected by sensors is initially aggregated and transmitted through industrial Ethernet to ensure the stability and real-time nature of data transmission. For the remote transmission part, taking advantage of the high speed and low latency characteristics of the 5G network, the data is quickly transmitted to the system server. At the same time, a data encryption mechanism is established, and the AES-256 encryption algorithm is used to encrypt the transmitted data to ensure the security and integrity of the data during transmission, prevent data leakage and malicious tampering, and ensure the reliable operation of the system.

[0024] Furthermore, in the digital twin construction and simulation module, the digital twin model is updated using an incremental update strategy. When the operating parameters of the equipment change, the system first determines whether the parameter change amplitude exceeds the threshold δ. If |Δx|>δ, the model update is triggered. During the update process, only the affected local model is updated and calculated. The formula is: M new = M old +ΔM, where M new is the updated digital twin model, M old is the model before update, and ΔM is the model variable calculated according to the parameter change. The calculation of ΔM is based on the finite element analysis method. According to the influence of the parameter change on the equipment structure and operation logic, the changes in the geometric shape and physical properties of the model are calculated. This strategy avoids the recalculation of the entire model, greatly improves the model update efficiency, and ensures that the digital twin model can quickly and accurately reflect the actual state of the equipment.

[0025] Furthermore, in the intelligent fault diagnosis module, a fault knowledge graph is established. By collecting multi-source information such as historical fault data, equipment manuals, and expert experience, fault entities, attributes, and relationships are extracted using knowledge extraction techniques to construct a fault knowledge graph. When performing fault diagnosis, combined with real-time monitoring data, reasoning and matching are carried out in the knowledge graph, which can not only quickly locate the fault type and cause, but also provide maintenance suggestions and preventive measures based on similar fault cases. The update of the knowledge graph is automatically optimized by continuously learning new fault data and expert knowledge, continuously improving the accuracy and comprehensiveness of fault diagnosis, and providing more valuable reference information for operation and maintenance personnel.

[0026] Compared with the prior art, the intelligent system for preventing coal blockage and purging and assisting coal flow in the coal conveying system has the following beneficial effects:

[0027] First, through the digital twin construction and simulation module, the system builds a digital twin model based on three-dimensional modeling technology to accurately simulate the physical structure, equipment layout, and operation logic of the coal conveying system. Combining historical data, real-time monitoring data, and a preset coal blockage scenario simulation algorithm, virtual simulation of possible coal blockage scenarios is carried out, so as to predict the possibility, location, and influence range of coal blockage, and issue early warnings in a timely manner. This method comprehensively considers the physical characteristics of coal flow and the probability of coal blockage, more realistically simulates the coal blockage scenario, and significantly improves the accuracy and efficiency of coal blockage early warning.

[0028] Second, by equipping with a remote control and debugging module, operation and maintenance personnel can log in to the system through remote terminal devices such as computers and mobile phones, view the physical equipment status mapped in real time by the digital twin model and the coal blockage scenario of virtual simulation, and remotely adjust the pressure, time, and interval parameters of the purging system based on the actual situation to achieve remote control. At the same time, relying on the digital twin model, real-time monitoring data, and virtual simulation results, the intelligent fault diagnosis module uses artificial intelligence algorithms and fault diagnosis models to quickly locate and analyze the causes of faults in the coal conveying system, and generate a detailed fault diagnosis report. The combination of this remote control and intelligent fault diagnosis not only improves the operation and maintenance efficiency, but also greatly improves the accuracy of fault diagnosis, providing strong support for timely handling of problems.

[0029] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be learned from the practice of the present invention. Brief Description of the Drawings

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0031] Figure 1 It is a flowchart for the function realization of the intelligent system for preventing coal blockage, purging, and assisting coal flow in the coal conveying system;

[0032] Figure 2 It is a schematic structural diagram of the intelligent system for preventing coal blockage, purging, and assisting coal flow in the coal conveying system. Detailed implementation manners

[0033] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the drawings and preferred embodiments, detail the specific implementation manners, structures, features, and their effects of the present invention as follows.

[0034] Embodiment 1

[0035] A thermal power plant is located in the north, where the winter temperature is often below -10°C. Due to the relatively high moisture content in the coal quality, the coal conveying system frequently experiences coal blockages at the elbows of the coal conveying pipeline, the interfaces between the coal dropping pipe and the coal feeder, etc. In the past, each manual cleaning required the unit to be shut down for 2 - 3 hours, which not only led to a 15% decrease in power generation efficiency but also posed safety hazards. To solve this problem, the power plant introduced an intelligent system for preventing coal blockage, purging, and assisting coal flow in the coal conveying system, which realizes coal blockage warning and intelligent purging through the collaborative work of multiple modules.

[0036] High-precision pressure sensors, temperature sensors, and vibration sensors are respectively installed at the elbows where coal is easily blocked in the coal conveying pipeline 3, the bottom of the coal dropping pipe, and the coal feeder inlet. The sensors collect data at a frequency of 10 times per second, such as the pressure fluctuation of the coal flow in the pipeline, the downward trend of the coal temperature with the ambient temperature, the abnormal vibration frequency of the equipment, etc. The data is aggregated to the on-site control cabinet through the industrial Ethernet and then encrypted and transmitted to the plant server through the 5G network. The system adopts a multi-parameter fusion evaluation algorithm with dynamic weights, and the formula is: S = ∑w i (t) × x i where S is the comprehensive evaluation value of the equipment operation status, n is the number of types of collected parameters, x i is the normalized real-time value of the i-th parameter, and w i (t) is the dynamic weight of the i-th parameter at time t. Automatically compare the historical operation data with the real-time parameters. When the pressure fluctuation of a certain section of the pipeline exceeds the normal range by 20% and the vibration frequency is abnormal, it is initially determined as a sign of coal blockage risk.

[0037] The digital twin model built based on 3D modeling technology fully restores the 150-meter pipeline route and equipment layout of the coal conveying system from the raw coal bunker to the boiler. Whenever the sensor data is updated, the color of the corresponding pipeline in the model will change dynamically according to the coal flow rate - it shows blue under normal conditions, gradually turns yellow when the flow rate decreases, and flashes red during coal blockage warning. The system uses a hybrid model of fluid dynamics and probability prediction. First, a basic fluid dynamics model of coal flow in the coal conveying pipeline is constructed based on the Navier-Stokes equation to describe the distribution of physical quantities such as the velocity and pressure of the coal flow. The formula is: where ρ is the density of the coal flow, is the velocity vector of the coal flow, p is the pressure, μ is the dynamic viscosity, is the external force. Then, a probability correction factor P i is introduced to establish a coal blockage probability prediction model where P is the total probability of coal blockage occurring in a certain area, m is the number of factors affecting coal blockage, P j is the probability of coal blockage caused by the jth factor. Combining the historical coal blockage data in the past 3 winters, high-risk areas are simulated: when the model shows that the coal flow velocity at a certain elbow is lower than the threshold and the predicted coal blockage probability exceeds 70%, a red warning box will pop up on the interface and mark "Coal blockage may occur within the next 2 hours".

[0038] Operation and maintenance personnel log in to the system through an industrial computer in the centralized control room to view the operating status of each device in the digital twin model in real time. When receiving a coal blockage warning for a certain coal chute, the operator clicks the "Intelligent Optimization Blowing" button, and the system automatically starts a hybrid algorithm of particle swarm optimization and fuzzy logic. The formula is: First, a blowing effect evaluation function E = ω1×T + ω2×P + ω3×S is established, where T is the blowing time, P is the blowing pressure, S is the degree of coal blockage relief, and ω1, ω2, ω3 are weight coefficients. The particle swarm optimization algorithm is used to search for the optimal solution in the parameter space. According to the current coal quality humidity, pipeline temperature and other parameters, the optimal combination of blowing pressure and duration is calculated, and the adjusted parameters are transmitted wirelessly to the blowing control system. Six groups of blowing nozzles located on both sides of the pipeline are started in sequence according to the optimized interval, and the digital twin model synchronously displays the dynamic simulation effect of coal flow dredging during the blowing process.

[0039] When the vibration sensor data still does not return to normal after blowing, the system starts a deep learning fault classification algorithm to extract features from multi-source data (sensor real-time data, digital twin simulation results, historical fault cases). For example, if the convolutional neural network analysis finds periodic abnormal peaks in the vibration signal, combined with reasoning based on the fault knowledge graph, it is determined as "coal adhesion and accumulation on the inner wall of the coal chute", and the specific blockage location is marked with a red arrow in the 3D model. At the same time, a diagnostic report is generated, suggesting "increase the blowing frequency in this area and arrange for manual inspection of the inner wall".

[0040] When the operation and maintenance personnel wear AR glasses and enter the coal conveying workshop, a transparent digital twin model is superimposed on the real scene, and holographic tags of parameters such as pressure and temperature are floating in real time on the surface of the pipeline. When a certain tag turns red, the glasses will emit a "beep" alarm sound. During debugging, the operator can adjust the purging parameters by swiping gestures, and the AR interface will synchronously display the prediction of the impact of parameter modification on the coal flow. If you need to view the fault report, after the voice command "retrieve the diagnosis of the No. 3 coal dropping pipe", the 3D animation will demonstrate the cause of coal blockage and the recommended solution, such as "due to long-term low temperature, coal slime adheres to the pipe wall, it is recommended to increase the tracing temperature in winter".

[0041] Embodiment 2

[0042] The blast furnace raw material conveying system of a steel plant needs to convey iron ore (particle size 5-50mm) and pulverized coal at the same time. Due to the large difference in material particles, mixed blockages frequently occur at the coal dropping pipe and reciprocating feeder below the screening device. In the past, it was necessary to stop the machine and clean it 4-5 times per shift, each time taking 40 minutes, resulting in unstable blast furnace feed rate and about 8% fluctuation in molten iron output. After introducing the intelligent system, the whole-process coal blockage prevention and control is realized through the cooperation of multiple modules.

[0043] Deploy pressure sensors (detecting the accumulated pressure of materials), flow sensors (monitoring the material conveying rate) and vibration sensors (capturing abnormal vibrations of equipment) at the outlet of the screening device, the middle of the coal dropping pipe and the inlet of the feeder. For example, when the sieve holes of the screening device are blocked by large-particle ore, the value of the upstream pressure sensor will continuously rise by 15% within 5 minutes, and at the same time, the flow sensor shows that the material flow rate drops by 30%. After the data is preliminarily processed by the industrial Ethernet, it is encrypted and transmitted to the central control room through the 5G network. The system uses the dynamic weight evaluation algorithm to calculate the comprehensive state value of the pressure, flow and vibration parameters according to the weight of 4:3:3. When this value exceeds the warning threshold, an alarm is triggered.

[0044] Build a full-process digital twin model covering the raw material bin, screening device, coal dropping pipe and feeder. The materials in the model are dynamically displayed with different color particles (red for iron ore, gray for pulverized coal). The system focuses on simulating the "throat" part below the screening device based on the hybrid model of fluid dynamics and probability prediction. When the model shows that the cross-sectional area of the pipeline is reduced by 40% due to the accumulation of iron ore and the moisture content of pulverized coal exceeds 12%, the coal blockage probability prediction model calculates that the coal blockage probability in this area reaches 85%, and marks the risk area with orange grid lines in the model, and at the same time sends a mobile phone text message warning to the operator.

[0045] After receiving the early warning on the mobile phone APP, the on-duty personnel enter the "Intelligent Optimization of Purge Parameters" interface. The system automatically calls the particle swarm optimization and fuzzy logic hybrid algorithm. According to the current material ratio (iron ore accounts for 60% and coal powder accounts for 40%) and the ambient temperature (25°C), it calculates the combination scheme of "high-pressure purge (for ore) and then pulse purge (for coal powder)". The adjusted parameters are sent to the on-site purge device, and 10 groups of nozzles are operated in the order of "first the two sides and then the middle". The digital twin model displays the dynamic process of material particles being purged and dispersed in real time until the coal flow velocity in the model returns to the normal 1.2m / s.

[0046] If the coal feeder vibration is still abnormal after purging, the system will start the deep learning fault algorithm and perform convolution feature extraction on the sensor data (such as the vibration spectrum) and the digital twin simulation data (such as the material flow trajectory). For example, when the algorithm finds an abnormal frequency component of 100Hz in the vibration signal, combined with the case feature of "coal feeder scraper jam" in the fault knowledge graph, it will quickly locate that there is an ore block stuck in the scraper inside the coal feeder, and highlight the fault location in the three-dimensional model. The diagnostic report also gives the treatment suggestion of "stopping the coal feeder operation and checking the scraper and bottom liner".

[0047] Operation and maintenance personnel wear AR helmets to enter the raw material conveying workshop. The surface of the equipment in the real scene is covered with a translucent digital twin model. A real-time flow data dashboard is suspended above the screening device, and the temperature distribution on the outer wall of the coal drop pipe is displayed with thermal imaging effects (the temperature in the blocked coal area is usually 5-8°C lower than the normal area). During debugging, the operator uses a voice command to "start the coal feeder purge mode", and the AR interface will pop up a parameter adjustment slider. The purge pressure can be modified by sliding gestures. At the same time, the model demonstrates the impact of the purge airflow on the material in real time. When viewing the fault report, the AR helmet projects a three-dimensional anatomical diagram, and uses arrow animation to demonstrate the cause and effect relationship of "ore block stuck in the scraper → causing abnormal vibration of the coal feeder", helping maintenance personnel quickly understand the fault mechanism.

[0048] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. The intelligent system for preventing coal blockage, purging and flow assistance in the coal conveying system is characterized in that The system includes the following components: a perception and data processing module, a digital twin construction and simulation module, a remote control and debugging module, an intelligent fault diagnosis module, and a human-machine interaction module; The perception and data processing module: Deploy various sensors for pressure, flow, temperature, and vibration at key parts of the coal conveying pipeline, coal dropping pipe, and coal feeder in the coal conveying system to collect equipment operation parameters in real time. The collected data is transmitted to this module through the communication network, preprocessed through cleaning, filtering, and analysis, and the effective information is extracted and stored in the database. At the same time, based on preset rules and algorithms, the operation status of the equipment is initially evaluated to determine whether there are any abnormalities; The digital twin construction and simulation module: Based on 3D modeling technology, build a digital twin model that accurately simulates the physical structure, equipment layout, and operation logic of the coal conveying system, and interact with the perception and data processing module to map the equipment operation data to the model in real time. Using this model, combined with historical data, real-time monitoring data, and a preset coal blockage scenario simulation algorithm, virtual simulation of possible coal blockage scenarios is carried out to predict the possibility, location, and impact range of coal blockage, and warnings are issued in a timely manner; The remote control and debugging module: Maintenance personnel log in to the system through remote terminal devices such as computers and mobile phones to view the physical equipment status and virtual simulation coal blockage scenarios mapped in real time by the digital twin model. Based on the actual situation, remotely adjust the pressure, time, and interval parameters of the purging system, and this module transmits the adjusted parameters to the purging control system to achieve remote control; The intelligent fault diagnosis module: Relying on the digital twin model, real-time monitoring data, and virtual simulation results, use artificial intelligence algorithms and fault diagnosis models to diagnose faults in the coal conveying system. Once abnormal equipment operation or coal blockage is detected, quickly locate the fault location, analyze the cause, and generate a detailed fault diagnosis report, which is displayed through the system interface or sent to relevant personnel via text message or email; The human-machine interaction module: Provide an intuitive and friendly user interface. Maintenance personnel can use this to view the real-time operation status of the coal conveying system, the digital twin model, the virtual simulation results, and the fault diagnosis report information, and can also input operation instructions to achieve various controls of the system.

2. The intelligent system for preventing coal blockage, purging, and assisting flow in the coal conveying system according to claim 1, wherein In the perception and data processing module, the evaluation of the equipment operation status adopts a multi-parameter fusion evaluation algorithm based on dynamic weights. The formula of this algorithm is: S = ∑w i (t) × x i Where S is the comprehensive evaluation value of the device operation status, n is the number of types of acquisition parameters, and x i is the real-time value after normalization of the i-th parameter, and w i (t) is the dynamic weight of the i-th parameter at time t.

3. The intelligent system for preventing coal blockage and purging and assisting flow in the coal conveying system according to claim 1, wherein, In the digital twin construction and simulation module, the coal blockage scenario simulation algorithm adopts a hybrid model based on fluid dynamics and probability prediction. First, a basic fluid dynamics model of coal flow in the coal conveying pipeline is constructed based on the Navier-Stokes equation to describe the distribution of physical quantities such as the velocity and pressure of the coal flow. The formula is: where ρ is the density of the coal flow, is the velocity vector of the coal flow, p is the pressure, μ is the dynamic viscosity, is the external force. Then, a probability correction factor P i is introduced to establish a coal blockage probability prediction model where P is the total probability of coal blockage occurring in a certain area, m is the number of factors affecting coal blockage, and P j is the probability of coal blockage caused by the jth factor.

4. The intelligent system for preventing coal blockage, purging, and flow assistance in the coal conveying system according to claim 1, wherein In the remote control and debugging module, there is an intelligent optimization sub-module for purging parameters. This sub-module uses a hybrid optimization algorithm based on particle swarm optimization and fuzzy logic to determine the purging parameters. First, an evaluation function for purging effect is established as E = ω1×T + ω2×P + ω3×S, where T is the purging time, P is the purging pressure, S is the degree of coal blockage alleviation, and ω1, ω2, ω3 are weight coefficients. The particle swarm optimization algorithm is used to search for the optimal solution in the parameter space. The particle position update formula is: x id (t + 1) = x id (t) + v id (t + 1), v id (t + 1) = w × v id (t) + c1r1(t)[p id - x id (t)] + c2r2(t)[p gd - x id (t)], where x id is the position of particle i in the d-th dimension, v id is the velocity of particle i in the d-th dimension, w is the inertia weight, c1, c2 are learning factors, r1, r2 are random numbers between [0, 1], p id is the individual optimal position of particle i, p gd is the global optimal position.

5. The intelligent system for preventing coal blockage, purging, and flow assistance in a coal conveying system according to claim 1, wherein In the intelligent fault diagnosis module, a fault classification and location algorithm based on deep learning is adopted to build a convolutional neural network model. The input layer receives multi-source data such as the preprocessed data from the perception and data processing module and the simulation data from the digital twin construction and simulation module. The convolutional layer extracts data features through convolutional kernels. The formula is: Among them is the j-th feature map of the l-th layer, M j is the set of input feature maps, is the i-th feature map of the (l - 1)-th layer, is the convolution kernel connecting the i-th input feature map and the j-th output feature map of the l-th layer, is the bias, f is the activation function, the pooling layer performs dimensionality reduction on the feature maps, the fully connected layer integrates the extracted features, and the output layer classifies the fault types through the softmax function. The formula is: where σ(z) j is the probability of the j-th type of fault, and z j is the j-th element of the input vector z, and K is the total number of fault types.

6. The intelligent system for preventing coal blockage, purging, and flow assistance in the coal conveying system according to claim 1, wherein The human-computer interaction module adopts augmented reality technology. Through wearing AR devices, operation and maintenance personnel can intuitively view the superimposed display of the digital twin model of the coal conveying system in the real scene. The operation parameters and fault warning information content of the equipment are marked in real time in the model. When performing remote debugging operations, the AR interface can interact through gesture recognition and voice commands. At the same time, the AR technology also supports generating three-dimensional visual fault diagnosis reports, presenting the fault location and cause information in the form of a three-dimensional model and animation demonstration.

7. The intelligent system for preventing coal blockage, purging, and flow assistance in the coal conveying system according to claim 1, wherein In the perception and data processing module, data communication adopts a hybrid communication method combining 5G and industrial Ethernet. At the equipment site, the data collected by sensors is initially aggregated and transmitted through industrial Ethernet. For the remote transmission part, taking advantage of the high-speed and low-latency characteristics of the 5G network, the data is quickly transmitted to the system server. At the same time, a data encryption mechanism is established, and the AES-256 encryption algorithm is used to encrypt the transmitted data.

8. The intelligent system for preventing coal blockage, purging, and flow assistance in the coal conveying system according to claim 1, wherein In the digital twin construction and simulation module, the digital twin model is updated using an incremental update strategy. When the device operating parameters change, the system first determines whether the parameter change amplitude exceeds the threshold δ. If |Δx| > δ, the model update is triggered. During the update process, only the affected local model is updated and calculated. The formula is: M new = M old + ΔM, where M new is the updated digital twin model, M old is the model before the update, ΔM is the model variable calculated based on the parameter change, and the calculation of ΔM is based on the finite element analysis method. According to the impact of the parameter change on the device structure and operating logic, the changes in the geometric shape and physical properties of the model are calculated.

9. The intelligent system for preventing coal blockage, purging, and flow assistance in the coal conveying system according to claim 1, characterized in that, In the intelligent fault diagnosis module, a fault knowledge graph is established. By collecting multi-source information such as historical fault data, equipment manuals, and expert experience, fault entities, attributes, and relationships are extracted using knowledge extraction technology to construct a fault knowledge graph. When performing fault diagnosis, combined with real-time monitoring data, reasoning and matching are carried out in the knowledge graph. The update of the knowledge graph is automatically optimized by continuously learning new fault data and expert knowledge, continuously improving the accuracy and comprehensiveness of fault diagnosis.

Images