Coal mill operation safety online monitoring and early warning system and method

Abstract

The invention relates to the field of coal mills, in particular to a coal mill operation safety online monitoring and early warning system and method.The coal mill operation safety online monitoring and early warning system is characterized in that a housing is integrally formed at the upper end of an upper shell; the upper end face of the housing is connected with a discharging pipe used for sending out ground coal powder and a second material pipe used for adding unground coal briquettes, a fan blade assembly is arranged on the outer side of the second material pipe and comprises a rotatable connecting base connected with the outer wall of the second material pipe, and a vertically-through collecting groove is formed in the surface of the connecting base. And the data acquisition module comprises a sensing module III arranged in the connecting seat. Through the fan blade assembly and an auxiliary structure, a data acquisition mode of dynamic rotation at the discharge port is realized, and data closest to a real working condition can be detected, so that the accuracy of the data input into the early warning management and control system is enhanced, and the possibility that the early warning management and control system executes a safety action mistakenly is reduced.

Description

Technical Field

[0001] The present invention relates to the field of coal mills, and in particular to an online monitoring and early warning system and method for coal mill operation safety. Background Art

[0002] When a thermal power plant is working, it needs to add pulverized coal for combustion. The burning pulverized coal is usually ground by a coal mill. The coal blocks placed inside the coal mill are ground by grinding rollers. After grinding, the air inlet at the bottom performs primary blowing, secondary blowing and other operations to blow the coal powder up from the pipe, and then passes through the inclined air duct in a swirling manner to form a centrifugal separation function, so that the coal powder with smaller particles is blown into the combustion chamber for combustion. However, during the operation of the coal mill, due to the presence of a lot of dust and the high operating temperature of the coal mill, and the fact that coal powder is a flammable and explosive substance, it is also necessary to equip the coal mill with a safety management and control system during operation to perform safety management work.

[0003] Conventional coal mill safety management and control systems use various sensors to collect different data at different locations, and then use the early warning management and control system to control the corresponding equipment to perform safety actions or issue early warning prompts to avoid danger. However, the installation of its sensors is usually fixed. Since the discharge airflow at the coal mill discharge port is in a turbulent state, the data detected by the fixed sensors are specific parameters at a specific location. The detected data is relatively one-sided, which may cause the parameters obtained by the early warning management and control system to be inaccurate when performing safety actions, thereby affecting the final execution effect. Summary of the Invention

[0004] The present invention aims to solve the problems existing in the prior art and provides the following technical solutions:

[0005] The coal mill operation safety online monitoring and early warning system includes: a lower shell for separating the fed coal blocks and the ground coal powder; an upper shell arranged above the lower shell; a data acquisition module and an early warning control system for executing early warnings and / or safety actions based on the data collected by the data acquisition module.

[0006] Specifically, a cover shell is integrally formed at the upper end portion of the upper shell, and the upper end surface of the cover shell is connected to a discharge pipe for sending out ground coal powder and a material pipe 2 for adding unground coal blocks. A fan blade assembly is provided on the outside of the material pipe 2, and the fan blade assembly includes a rotatable connecting seat connected to the outer wall of the material pipe 2. A vertically penetrating collection groove is provided on the surface of the connecting seat. The data acquisition module includes a sensing module 3 arranged inside the connecting seat. When the discharge pipe discharges material, the internal airflow drives the fan blade assembly to drive the collection groove to rotate to perform dynamic sampling of the sensing module 3.

[0007] As an improvement of the above technical solution, the lower shell includes an outer shell connected to the lower end of the upper shell, an inner shell located inside the upper shell, and a material pipe. The inner shell is fixed to the inner wall of the upper shell by a penetrating fixing bolt, and the material pipe one is connected to the inner wall of the cover shell by a penetrating fixing bolt, and the material pipe one is connected to the material pipe two.

[0008] As an improvement of the above technical solution, the connecting seat is arranged between material pipe one and material pipe two, and the upper and lower end surfaces of the connecting seat are respectively fitted with material pipe one and material pipe two, and a spherical groove is provided at the fitting surface, and a ball is provided in the spherical groove. A groove for installing sensing module three is provided inside the connecting seat, and a cover plate is provided on the upper end surface of the groove.

[0009] As an improvement of the above technical solution, the data acquisition module includes a sensing module 1 and a sensing module 2. The outer side of the lower shell is provided with an air inlet and a side frame. The sensing module 2 is arranged on the inner wall of the air inlet, and the sensing module 1 is arranged inside the side frame. The air inlet is used to blow air into the interior to drive the dust after grinding to rise. The side frame is used to connect the roller frame and the hydraulic cylinder for adjusting the height of the roller frame.

[0010] As an improvement of the above technical solution, a folding assembly is provided inside the upper shell, and the folding assembly is located in a cavity between the upper inner side of the inner shell and the lower outer side of the cover shell.

[0011] As an improvement of the above technical solution, the early warning and control system includes a data processing module, a causal reasoning module, a safety control module, and a feedback learning module. The data processing module is used to preprocess the data collected by the data acquisition module. The causal reasoning module is used to infer the executable safety actions of the faulty equipment based on the correlation characteristics between the data. The safety control module is used to calculate and obtain the execution parameters of the executable safety actions based on the data collected by the data processing module. The feedback learning module is used to output alarm feedback when the executable safety action cannot be analyzed, and record the operator's execution actions and feed them back to the causal reasoning module.

[0012] The coal mill operation safety online monitoring and early warning method is applied to the coal mill operation safety online monitoring and early warning system as described in any one of the above technical solutions, comprising the following steps:

[0013] S10: Preprocessing and feature extraction are performed on the data collected by the data collection module through the data processing module.

[0014] S20: Input the extracted features into the causal reasoning module to construct a triple causal chain, and construct the triple causal chain into a data chain with a unique pointer, which points to a safe action without parameters.

[0015] S30: The security control module obtains execution parameters of the security action according to the features extracted by the data processing module and executes the security action.

[0016] S40: If the data chain points to any safety action, the early warning action is initiated, and data recording is started to record the status of manual repair or each action feedback and feed it back to the causal reasoning module.

[0017] As an improvement to the above technical solution, outliers are removed, data is aligned, noise is filtered, and normalization is performed.

[0018] As an improvement of the above technical solution, the triple causal chain is in the form of: detection parameters -> equipment structure -> fault problem. The parameters in each tuple are linked to form a data chain with a unique pointer. The data chain points to a safety action without parameters, and the safety action is recorded in the knowledge graph.

[0019] As an improvement of the above technical solution, the safety control module includes at least a static mapping table and a fuzzy logic controller. The static mapping table is used to respond to safety actions and output preset control parameters. The fuzzy logic controller is used to receive features extracted by the data processing module, perform fuzzy reasoning based on multi-variable coupling relationships, and output specific control parameters.

[0020] Beneficial effects of the present invention:

[0021] Through the provision of the fan blade assembly and the auxiliary structure, a dynamic rotation data collection method is realized at the discharge port. Since the discharge pipe is connected to the space where the fan blade assembly is located, when the entire coal mill is working, the airflow flows through the fan blade assembly, and the airflow will pass through the collection slot on the connecting seat. The airflow and the coal powder particles mixed in the airflow will pass through the internally provided sensing module three, so that the data closest to the actual working conditions can be detected, thereby enhancing the accuracy of the data input into the early warning control system and reducing the possibility of the early warning control system performing safety actions incorrectly. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a three-dimensional structural diagram of the coal mill operation safety online monitoring and early warning system of the present invention;

[0023] Figure 2 for Figure 1 A magnified structural diagram at center A;

[0024] Figure 3 A side view of the coal mill operation safety online monitoring and early warning system of the present invention;

[0025] Figure 4 for Figure 3 Isometric cross-sectional view at aa in the middle;

[0026] Figure 5 for Figure 4 The enlarged structure diagram at B in the middle;

[0027] Figure 6 This is an explosion structure diagram of the coal mill operation safety online monitoring and early warning system of the present invention;

[0028] Figure 7 for Figure 6 The enlarged structure diagram at C in the middle;

[0029] Figure 8 for Figure 6 The enlarged structure diagram at D in the middle;

[0030] Figure 9 This is a diagram showing the connection status of the coal mill operation safety online monitoring and early warning system and the coal mill body before and after installation;

[0031] Figure 10 This is a logic block diagram of the early warning control system of the present invention;

[0032] Figure 11 This is a flow chart of the coal mill operation safety online monitoring and early warning method of the present invention.

[0033] Reference numerals:

[0034] 10. Lower shell; 11. Outer shell; 12. Inner shell; 13. Material pipe 1; 14. Air inlet; 15. Side frame; 151. Mounting slot; 152. Snap-fit ​​plate; 16. Fixing bolt;

[0035] 20. Upper shell; 21. Folding assembly; 211. Adjusting ring; 2111. Protrusion; 2112. Connecting portion 1; 212. Telescopic rod; 2121. Rod sleeve; 2122. Shaft; 2123. Threaded portion; 2124. Connecting portion 2; 213. Mounting bracket; 214. Rotating shaft; 2141. Steering plate; 2142. Folding plate;

[0036] 22. Cover; 221. Discharge pipe; 222. Second discharge pipe;

[0037] 23. Fan blade assembly; 231. Connecting seat; 232. Collection slot; 233. Cover plate; 234. Ball bearing;

[0038] 30. Data acquisition module; 31. Perception module 1; 32. Perception module 2; 33. Perception module 3;

[0039] 40. Early warning and control system; 41. Data processing module; 42. Causal reasoning module; 43. Safety control module; 44. Feedback learning module. DETAILED DESCRIPTION

[0040] The following describes the embodiments of the present invention through specific examples. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention.

[0041] Conventional coal mill safety management and control systems use various sensors to collect different data at different locations, and then use the early warning management and control system to control the corresponding equipment to perform safety actions or issue early warning prompts to avoid danger. However, the installation of its sensors is usually fixed. Since the discharge airflow at the coal mill discharge port is in a turbulent state, the data detected by the fixed sensors are specific parameters at a specific location. The detected data is relatively one-sided, which may cause the parameters obtained by the early warning management and control system to be inaccurate when performing safety actions, thereby affecting the final execution effect.

[0042] In order to solve the above problems, the following embodiments are provided:

[0043] Example 1

[0044] To solve the above problem, please refer to Figures 1 to 10 , provides an online monitoring and early warning system for coal mill operation safety, including: a lower shell 10 for separating fed coal blocks and ground coal powder; an upper shell 20 arranged above the lower shell 10; a data acquisition module 30 and an early warning control system 40 for executing early warnings and / or safety actions based on data collected by the data acquisition module 30.

[0045] Specifically, a cover shell 22 is integrally formed at the upper end of the upper shell 20, and the upper end surface of the cover shell 22 is connected to a discharge pipe 221 for sending out the ground coal powder and a second material pipe 222 for adding unground coal blocks. A fan blade assembly 23 is provided on the outside of the second material pipe 222. The fan blade assembly 23 includes a rotatable connecting seat 231 connected to the outer wall of the second material pipe 222. A vertically penetrating collection groove 232 is provided on the surface of the connecting seat 231. The data acquisition module 30 includes a sensing module three 33 arranged inside the connecting seat 231. When the discharge pipe 221 discharges material, the internal airflow drives the fan blade assembly 23 to drive the collection groove 232 to rotate to perform dynamic sampling of the sensing module three 33.

[0046] The connecting seat 231 as a connecting part can rotate itself, and its power source comes from the fan blades connected to it. When the air flow blows through, the fan blades rotate under the action of external force, and the connecting seat 231 fixed to the fan blades will also produce synchronous turning, and rotate at different speeds due to the flow rate of the air flow.

[0047] The equipment structure of this system needs to rely on the main structure of the coal mill to operate, so first of all, the entire lower shell 10 needs to be installed above the grinding roller on the coal mill. The installation form is as follows: Figure 9 As shown, when working, the coal powder will be blown out from the discharge pipe 221 along with the air flow, and during the blowing process, the air flow will drive the fan blade assembly 23 to rotate, and the connecting seat 231, as a part of the fan blade assembly 23, will rotate along with the fan blade. In this case, the collection slot 232 opened on the connecting seat 231 will rotate along with it. Since the air flow entering the interior of the cover 22 is in a spiral upward operation state, the air flow flowing through the rotating collection slot 232 is dynamic. In this case, the dynamic detection sampling function can be realized, and the data collected in this way is also dynamic. When the early warning control system 40 receives this dynamic sampling data for verification calculation, the possibility of misjudgment of the obtained result is lower, and the accuracy of control and early warning is higher.

[0048] In order to further improve the structural design of coal powder separation, please refer to Figures 1 to 4 The lower shell 10 includes an outer shell 11 connected to the lower end of the upper shell 20, an inner shell 12 located inside the upper shell 20, and a material pipe 13. The inner shell 12 is fixed to the inner wall of the upper shell 20 by a penetrating fixing bolt 16. The material pipe 13 is connected to the inner wall of the cover shell 22 by a penetrating fixing bolt 16. The material pipe 13 is connected to the material pipe 2 222.

[0049] This part of the structure is the design of the separation flow channel inside the existing coal mill, so only its structural features are explained in this embodiment. Its principle is to form multiple groups of connected separation cavities inside through the inner shell 12 and the cover shell 22 and other structures of the inner layer, and to form a vortex centrifugal state inside between the cover shell 22 and the inner shell 12 through the gas flow channel with an inclined angle, thereby separating the coal powder with smaller particles from the coal powder with larger particles, and outputting the coal powder with smaller particles from the material pipe 222.

[0050] In order to further improve the structural design of the swirl, please refer to Figure 1 、 Figure 2 、 Figure 6 as well as Figure 7 Specifically, the interior of the upper shell 20 is provided with a folding assembly 21, and the folding assembly 21 is located in a cavity between the upper inner side of the inner shell 12 and the lower outer side of the cover shell 22.

[0051] The hinge structure forms an airflow restriction, and the airflow flows between two adjacent hinges. When the hinge is tilted, the angle of the airflow will also change. When the hinge is designed to be annular, an annular swirl state is formed, realizing the function of centrifugal separation. Specifically, the structural design of the hinge assembly 21 is as follows:

[0052] The telescopic rod 212 comprises an adjusting ring 211, a telescopic rod 212, a mounting bracket 213 and a rotating shaft 214. The adjusting ring 211 is provided with a penetrating protrusion 2111, and a connecting portion 1 2112 is provided on the outer side of the adjusting ring 211. The telescopic rod 212 comprises a rod sleeve 2121 fixed on the mounting bracket 213. The rod sleeve 2121 and the mounting bracket 213 are connected by an axial body. The rod sleeve 2121 is penetrated by a shaft rod 2122. The shaft rod 2122 is inserted into the rod sleeve 2121. A threaded portion 2123 is provided on one side of the rod sleeve 2121. The threaded portion 2123 is threadedly connected to the rod sleeve 2121. The other end of the shaft rod 2122 is provided with a connecting portion 2124. The connecting portion 2124 is connected to the connecting portion 1 2112 by an axial body. When it is long, the adjusting ring 211 will rotate. Further, the rotating shaft 214 will pass through the interior of the upper shell 20 and be connected to the folding plate 2142, and the other end of the shaft rod 2122 is located outside the upper shell 20 and is connected to the steering plate 2141. A movable rail groove is provided on the surface of the steering plate 2141, and the protrusion 2111 is inserted into the interior of the movable rail groove. In this way, the rotating adjusting ring 211 will drive the protrusion 2111 to rotate, and the rotating protrusion 2111 will move inside the movable rail groove, thereby causing the shaft rod 2122 to produce an axial rotation state, thereby forming the angle adjustment of the bottom folding plate 2142. As long as the angles of all folding plates 2142 are consistent, a stable swirl state can be output, thereby realizing the function of centrifugal separation of coal powder particles.

[0053] In order to ensure the rotatable structure design of the connecting seat 231, please refer to Figures 1 to 8 The connecting seat 231 is arranged between the material pipe 13 and the material pipe 2 222. The upper and lower end surfaces of the connecting seat 231 are respectively fitted with the material pipe 13 and the material pipe 2 222. A spherical groove is provided at the fitting surface, and a ball 234 is provided in the spherical groove. A groove for installing the sensing module 3 33 is provided inside the connecting seat 231, and a cover plate 233 is provided on the upper end surface of the groove.

[0054] The original integrated material pipe is divided into two parts. The upper material pipe 222 is integrally formed with the upper shell 20, while the lower material pipe 222 is fixed by a fixing bolt 16. This method ensures the airflow without affecting the airflow direction of the swirl, and can also ensure the detachability of the structure. Through this design, some space can be left in the middle, and the connecting seat 231 is embedded in this position. While the ball bearing 234 is used to ensure the rotation, an arc-shaped baffle can be set on the inner wall of the material pipe 222 to cover the connection to prevent the falling coal blocks from affecting the coal powder particle size inside the cover 22. Figure 5 shown.

[0055] In order to further improve the design of the data acquisition module 30, please refer to Figures 1 to 8The data acquisition module 30 includes a sensing module 1 31 and a sensing module 2 32. The outer side of the lower shell 10 is provided with an air inlet 14 and a side frame 15. The sensing module 2 32 is arranged on the inner wall of the air inlet 14, and the sensing module 1 31 is arranged inside the side frame 15. The air inlet 14 is used to blow air into the interior to drive the dust after grinding to rise. The side frame 15 is used to connect the roller frame and the hydraulic cylinder for adjusting the height of the roller frame.

[0056] Side frame 15 as Figure 7 As shown, the side frame 15 is further provided with a mounting groove 151, which is used to engage with the end of the roller frame. The end of the roller frame is the part connected to the hydraulic cylinder, which is usually fixed to the roller frame by the rod of the hydraulic cylinder. The fixed connection passes through the mounting groove 151, and the other end face of the side frame 15 is a snap-fit ​​plate 152 fixed by screws. The vibration sensor in the sensing module 31 can be set on the inner surface of the snap-fit ​​plate 152 to transmit the vibration of the entire equipment shell through the side frame 15.

[0057] The sensing module 31 is usually a vibration sensor, a flame detector, and a temperature sensor. The vibration sensor is used to sense the vibration of the equipment shell and can analyze whether the equipment is in normal working condition based on the vibration spectrum of the equipment shell. The flame detector is used to detect whether sparks appear inside the entire system shell. The flame detector can usually be installed at a fixed position between the rod and the roller frame to detect the internal space, and the temperature sensor senses the temperature inside the system shell.

[0058] The sensing module 2 32 is usually a wind speed sensor, a pressure sensor, and a temperature sensor. The wind speed sensor is used to monitor the wind speed of the air flow sent into the air inlet 14, and the pressure sensor is used to detect the wind pressure at the air outlet. If the pressure changes, there may be problems such as gas leakage. The temperature sensor is used to monitor the temperature of the air inlet 14 to avoid internal combustion caused by excessive temperature.

[0059] Perception module three 33 is usually a CO gas detection sensor, an oxygen concentration sensor, and a laser scattering particle sensor. These three sets of sensors are used to detect the airflow and material particles at the discharge port. The CO gas detection sensor and the oxygen concentration sensor can detect the CO gas and oxygen concentration in the last airflow, so as to analyze whether there will be safety risks such as combustion and explosion under the current working conditions. The laser scattering particle sensor is used to monitor the particle size inside the passing airflow. In this way, the actual effect of the current particle separation can be judged to avoid the separation particle size being different from the actual demand.

[0060] The sampling nodes of the above-mentioned sensors all adopt a dynamic frequency sampling method. In this embodiment, the sampling frequency is accelerated according to the number of abnormal point data detected. The minimum detection frequency for normal operation without abnormal points is usually 1 Hz-100 Hz, which is specifically determined according to the use environment of different sensors. For example, the sampling of vibration sensors requires high frequency, so it may be 100 Hz for detection under normal circumstances, but when abnormal point data is detected, its detection frequency may reach 1 khz-10 khz, while gas sensors, pressure sensors, etc. only need lower frequencies for detection.

[0061] In one embodiment, see Figure 10 The early warning control system 40 includes a data processing module 41, a causal reasoning module 42, a safety control module 43, and a feedback learning module 44.

[0062] Data is collected by the data acquisition module 30. The collected data is pre-processed by the data processing module 41, typically by data cleaning and feature extraction. After processing, the extracted features are input into the causal reasoning module 42. Based on the correlation features between the data, the executable safety action for the faulty device is inferred. The executable safety action does not contain actual execution parameters, but only includes a response strategy for the fault. For example, in the case of a load abnormality, the data acquisition module 30 first detects the load abnormality parameters. The causal reasoning module 42 then uses the load abnormality parameters to query the executable action, which is load reduction. Here, load reduction is simply an execution action without specific reduction parameters. The safety control module 43 calculates and analyzes the execution parameters of the executable safety action. Finally, the feedback learning module 44 records the specific situation of the executable safety action and the state change of the device. In addition, if the executable safety action cannot be found, an early warning feedback is output, and after waiting for the operator to inspect and repair, the operator's execution action is recorded and fed back to the causal reasoning module 42 to optimize the system design.

[0063] Example 2

[0064] To further understand how the entire system works, see Figure 10 and Figure 11 A method for online monitoring and early warning of coal mill operation safety is provided, which is applied to the online monitoring and early warning system for coal mill operation safety as described in the first embodiment, and is characterized by comprising the following steps:

[0065] S10: The data processing module 41 performs preprocessing and feature extraction on the data collected by the data collection module 30 .

[0066] Preprocessing includes at least: removing outliers, data alignment, noise filtering and standardization.

[0067] For the removal of outliers, the 3σ principle, IQR method and other methods can usually be used to identify and eliminate obviously erroneous data. Data alignment is achieved by aligning the timestamps and unifying the sampling rates of data of different frequencies collected by various types of sensors. Noise filtering can usually use low-pass filters, sliding averages, wavelet denoising and other methods to process signal noise. Finally, through standardization, data of different dimensions are standardized to the same range to facilitate subsequent data processing.

[0068] Feature extraction typically involves extracting time- and frequency-domain features. Time-domain features typically involve directly detecting signal changes and obtaining energy-related parameters such as RMS, absolute mean, and crest factor. Frequency-domain features typically involve converting the time-domain signal into a frequency state through an FFT transform, and then extracting the time- and frequency-domain signals through wavelet transforms or short-time Fourier transforms. After feature extraction is complete, step S20 is executed.

[0069] S20: Input the extracted features into the causal reasoning module 42 to construct a triple causal chain, and construct the triple causal chain into a data chain with a unique pointer, which points to a safe action without parameters.

[0070] Unlike traditional data judgment, coal mills have fewer devices, and the correlation between the internal power structure and the parameters that need to be detected is not too complicated. Therefore, when reasoning about causal relationships, data, equipment, and problems can be associated into a triple. Unlike traditional triples, the triple constructed by the coal mill can be simplified into a single pointing chain. Specifically, the triple causal chain is in the form of: detection parameters -> equipment structure -> fault problem. The parameters in each tuple are linked to form a data chain with a unique pointing direction. The data chain points to a safety action without parameters, and the safety action is recorded in the knowledge graph.

[0071] The construction of the knowledge graph is a one-way graph with pointing arrows formed based on the detection parameters, the device structure corresponding to the detection parameters, and the problems generated by the device results corresponding to the detection parameters. The pointing arrows between each two tuples are single. Here, when identifying the detection parameters, the specific values ​​of the parameters are not identified, but the types of the parameters are identified to determine which part of the device structure where the parameters appear. The detection parameters in the problems generated by the device results corresponding to the detection parameters refer to the parameters detected by the sensor. Specifically, they are distinguished through two aspects. One is the problem pointed to by a single parameter, for example, temperature parameter -> air inlet blower -> low wind speed leads to high temperature. The second is the problem pointed to by multiple parameters, for example, temperature parameter, voltage parameter -> a certain device -> excessive load leads to excessive heat. The specific execution parameters of the safety action are determined and calculated in the safety control module 43. This step-by-step processing method can simplify the complexity of the calculation, thereby speeding up the calculation.

[0072] S30: The security control module 43 obtains execution parameters of the security action according to the features extracted by the data processing module 41 and executes the security action.

[0073] Specifically, the safety control module 43 includes at least a static mapping table and a fuzzy logic controller. The static mapping table is used to respond to safety actions and output preset control parameters. The fuzzy logic controller is used to receive the features extracted by the data processing module 41, perform fuzzy reasoning based on the multi-variable coupling relationship, and output specific control parameters.

[0074] Static mapping tables are suitable for direct modification of some simple parameters. These mapping tables are usually a collection of historical data. In the historical data, all safety actions, whether manually performed or automatically executed by the system, will be recorded. This recorded information can not only be used for learning, but also generate a static mapping table to match the data with the safety actions. When the same problem occurs again in the future, the data in the static mapping table can be directly called for adjustment, which is more efficient.

[0075] A fuzzy logic controller typically defines fuzzy sets and membership functions for each input variable. Then, based on experience or historical data analysis, it specifies "if-then" control rules. For example, "If the temperature is high and vibration is severe, the load should be significantly reduced." Fuzzy inference algorithms are used for calculations, and defuzzification methods such as the center of gravity method are used to output specific control parameters. However, since the specific safety action has already been determined in step S20, the fuzzy logic controller's calculations omit the step of analyzing and specifying "if-then" control rules, thereby improving the efficiency of the entire analysis process.

[0076] S40: If the data link points to any safety action, the early warning action is initiated, and data recording is started to record the status of manual repair or each action feedback and feed it back to the causal reasoning module 42.

[0077] The execution of step S40 is through the feedback learning module 44. In the aforementioned scheme, the analysis and generation of specific data for the execution action have been completed. Therefore, it is only necessary to adjust the equipment according to the parameters. However, if it is found in step S20 that the location data pointed to by the data link is empty, in this case it is considered that an undiscovered fault has occurred, and the staff is reminded to make modifications by issuing an early warning alarm. In this process, predictions can usually be made based on historical data. It can usually be considered to be combined with the specially designed knowledge graph used in this embodiment for analysis. This can provide faster association identification and generate prediction solutions. This part can add and call the corresponding functional modules according to the actual needs of the user.

[0078] After manual modification, the parameter changes of the corresponding equipment and modules during the modification period will be recorded, and the safety actions pointed to by the corresponding triplet causal chain will be recorded. This ensures that a quick response can be made when the problem occurs again next time. This design also ensures that the equipment will become more and more perfect with long-term application. Of course, it will also lead to an increase in data storage volume. It is possible to consider designing an independent database for separate storage to avoid data confusion.

[0079] In addition to manual modification, the automatic modification method using the method of this embodiment will also trigger data recording. The data records here are used for comparison to analyze whether the same operation will produce different data results. This is because different results may appear when the same operation is applied on the device. In order to avoid the system mistakenly believing that the system security management has been completed, data verification is also required to ensure that the device is in a safe and stable working state after the data is modified. If there is a slight fluctuation that does not affect the operation of the device, it will be recorded and marked. If it still appears for a long time in subsequent operations, it will be marked and recorded in the log and fed back to the operator. If there is a large fluctuation, an early warning will be issued and fed back to the operator for processing.

[0080] The above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the same. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. The coal mill operation safety online monitoring and early warning system is characterized by: include: A lower shell (10) for separating fed coal blocks and ground coal powder; An upper shell (20) is arranged above the lower shell (10), and a cover shell (22) is integrally formed at the upper end of the upper shell (20). The upper end surface of the cover shell (22) is connected to a discharge pipe (221) for sending out ground coal powder and a second material pipe (222) for adding unground coal blocks. A fan blade assembly (23) is provided on the outer side of the second material pipe (222). The fan blade assembly (23) includes a rotatable connecting seat (231) connected to the outer wall of the second material pipe (222). A vertically penetrating collecting groove (232) is provided on the surface of the connecting seat (231); A data acquisition module (30) and an early warning control system (40) for executing early warnings and / or safety actions based on data collected by the data acquisition module (30), wherein the data acquisition module (30) includes a sensing module three (33) arranged inside a connecting seat (231), and when the discharge pipe (221) discharges material, the internal airflow drives the fan blade assembly (23) to drive the collection slot (232) to rotate to perform dynamic sampling of the sensing module three (33).

2. The coal mill operation safety online monitoring and early warning system according to claim 1 is characterized by: The lower shell (10) includes an outer shell (11) connected to the lower end of the upper shell (20), an inner shell (12) located inside the upper shell (20), and a material pipe (13). The inner shell (12) is fixed to the inner wall of the upper shell (20) by a fixed bolt (16) inserted therethrough. The material pipe (13) is connected to the inner wall of the cover shell (22) by a fixed bolt (16) inserted therethrough. The material pipe (13) is communicated with the material pipe (222).

3. The coal mill operation safety online monitoring and early warning system according to claim 2 is characterized by: The connecting seat (231) is arranged between the material pipe (13) and the material pipe (222), and the upper and lower end surfaces of the connecting seat (231) are respectively fitted with the material pipe (13) and the material pipe (222), and a spherical groove is provided at the fitting surface, and a ball (234) is provided in the spherical groove. A groove for installing the sensing module (33) is provided inside the connecting seat (231), and a cover plate (233) is provided on the upper end surface of the groove.

4. The coal mill operation safety online monitoring and early warning system according to claim 1 is characterized by: The data acquisition module (30) includes a sensing module 1 (31) and a sensing module 2 (32). An air inlet (14) and a side frame (15) are provided on the outside of the lower shell (10). The sensing module 2 (32) is provided on the inner wall of the air inlet (14). The sensing module 1 (31) is provided inside the side frame (15). The air inlet (14) is used to blow air into the interior to drive the dust after grinding to rise. The side frame (15) is used to connect the roller frame and the hydraulic cylinder for adjusting the height of the roller frame.

5. The coal mill operation safety online monitoring and early warning system according to claim 2 is characterized by: A folding assembly (21) is provided inside the upper shell (20), and the folding assembly (21) is located in a cavity between the upper inner side of the inner shell (12) and the lower outer side of the cover shell (22).

6. The coal mill operation safety online monitoring and early warning system according to any one of claims 1 to 5, characterized in that: The early warning control system (40) includes a data processing module (41), a causal reasoning module (42), a safety control module (43), and a feedback learning module (44); The data processing module (41) is used to pre-process the data collected by the data collection module (30); the causal reasoning module (42) is used to infer the executable safety action of the faulty equipment based on the correlation characteristics between the data; the safety control module (43) is used to calculate and obtain the execution parameters of the executable safety action based on the data collected by the data processing module (41); the feedback learning module (44) is used to output alarm feedback when the executable safety action cannot be analyzed, and record the execution action of the operator and feed it back to the causal reasoning module (42).

7. A method for online monitoring and early warning of coal mill operation safety, applied to the online monitoring and early warning system for coal mill operation safety according to any one of claims 1 to 6, characterized in that: The steps include: S10: Preprocessing and feature extraction of the data collected by the data collection module (30) by the data processing module (41); S20: Input the extracted features into the causal reasoning module (42) to construct a triple causal chain, and construct the triple causal chain into a data chain with a unique pointer, which points to a safe action without parameters; S30: The security control module (43) obtains execution parameters of the security action based on the features extracted by the data processing module (41) and executes the security action; S40: If the data chain points to any safety action, the early warning action is started, and data recording is started to record the status of manual repair or each action feedback and feed it back to the causal reasoning module (42).

8. The coal mill operation safety online monitoring and early warning method according to claim 7 is characterized by: The preprocessing includes at least: removing outliers, data alignment, noise filtering and standardization.

9. The coal mill operation safety online monitoring and early warning method according to claim 7, characterized in that: The triple causal chain is in the form of: detection parameters -> equipment structure -> fault problem. The parameters in each tuple are linked to form a data chain with a unique pointer. The data chain points to a safety action without parameters, and the safety action is recorded in the knowledge graph.

10. The coal mill operation safety online monitoring and early warning method according to claim 7, characterized in that: The safety control module (43) at least includes a static mapping table and a fuzzy logic controller; The static mapping table is used to respond to safety actions and output preset control parameters. The fuzzy logic controller is used to receive features extracted by the data processing module (41), perform fuzzy reasoning based on multi-variable coupling relationships, and output specific control parameters.

Images