Self-adaptive ship unloading method based on coal hardness

Abstract

The invention discloses a self-adaptive ship unloading method based on coal hardness, which comprises the following steps: detecting the closing speed of a grab bucket driving mechanism of a ship unloader in real time, acquiring a grab bucket pressure value through an oil cylinder pressure sensor, and calculating real-time sensing data representing the coal hardness according to the grab bucket pressure value; inputting the real-time sensing data into a hardness-control parameter mapping relation table, and querying and outputting corresponding target operation parameters; by introducing variables such as environment temperature and moisture content, target operation parameters are corrected in real time by adopting a dynamic compensation mechanism, and the grab bucket type ship unloader is controlled to carry out subsequent grabbing operation according to the target operation parameters; the ship unloading efficiency in unit time is monitored, the ship unloading efficiency serves as a feedback signal, self-adaptive correction is conducted on the hardness-control parameter mapping relation table, the remarkable improvement of the ship unloading efficiency, the prolonging of the service life of equipment and the reduction of comprehensive energy consumption are achieved, and the purpose of intelligent operation is achieved.

Description

Technical Field

[0001] This invention relates to the field of port bulk cargo handling control technology, specifically to an adaptive unloading method based on coal hardness. Background Technology

[0002] In the bulk cargo handling sector, such as ports and thermal power plants, grab unloaders are core equipment for coal unloading operations. Their traditional operating mode heavily relies on operator experience. Key parameters such as unloading speed, grab depth, and closing torque are typically preset and fixed in the control system based on the average characteristics of a single coal type or conservative experience. However, the hardness of coal varies significantly in actual operations. For example, soft lignite and hard anthracite have vastly different physical properties. This significant difference poses a serious challenge to fixed operating parameters. The existing "one-size-fits-all" control method exposes a series of inherent defects when dealing with varying coal qualities, mainly in the following three aspects: (1) Low coal unloading efficiency: When unloading soft coal such as lignite, the grab bucket fails to exert its maximum capacity due to overly conservative parameter settings, resulting in insufficient single grab volume and wasting equipment throughput. According to statistics, throughput may decrease by more than 20% under soft coal conditions. When unloading hard coal such as anthracite, the equipment overload protection is frequently triggered and the machine stops due to excessive cutting resistance, and the operation process is repeatedly interrupted, which also seriously restricts the unloading efficiency. (2) Severe equipment wear: Under hard coal conditions, the fixed strong operation mode will cause structural components such as grab bucket and tooth tip to bear huge impact and stress, resulting in a significant increase in the deformation rate of grab bucket. At the same time, the transmission system such as wire rope and motor will work on the edge of overload for a long time, which will aggravate their fatigue damage and directly lead to a significant reduction in the service life of key components such as wire rope, increasing the maintenance cost and safety risk of the equipment. (3) Insufficient level of intelligence: Existing methods lack the ability to perceive and respond to the key working condition variable of coal quality in real time, and cannot realize the intelligent control of "unloading according to coal". The entire operation process is a rigid open-loop system, which cannot adaptively adjust the operation strategy according to the dynamic changes in coal hardness, so as to achieve a dynamic optimal balance between efficiency, energy consumption and equipment protection. Summary of the Invention

[0003] The purpose of this invention is to provide an adaptive unloading method based on coal hardness to solve the technical problems of low coal unloading efficiency, serious equipment wear and tear, and insufficient intelligence level in the existing coal unloading machine.

[0004] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution: This invention provides an adaptive unloading method based on coal hardness, comprising the following steps: The closing speed of the grab bucket drive mechanism of the ship unloader is detected in real time, and the grab bucket pressure value is collected through the hydraulic cylinder pressure sensor. The real-time sensing data characterizing the hardness of coal is calculated based on the grab bucket pressure value. The real-time sensing data is input into the hardness-control parameter mapping table, and the corresponding target operation parameters are queried and output. The target operation parameters include at least the grab bucket cutting speed, the closed bucket motor torque threshold, or the lifting speed. By introducing variables such as ambient temperature and moisture content, the target operating parameters are dynamically compensated in real time to correct them, and the grab unloader is controlled to perform subsequent grabbing operations with the target operating parameters. The unloading efficiency per unit time is monitored, and the unloading efficiency is used as a feedback signal to adaptively correct the hardness-control parameter mapping table.

[0005] As a preferred embodiment of the present invention, the closing speed of the unloader's grab bucket drive mechanism is detected in real time, and the grab bucket pressure value is collected through a hydraulic cylinder pressure sensor. Real-time sensing data characterizing coal hardness is calculated based on the grab bucket pressure value, including: The closing speed of the grab bucket drive mechanism of the ship unloader is monitored in real time in two stages: the grab bucket closing stage and the switching stage. During the grab closing phase, the peak value of the hydraulic cylinder pressure sensor is collected. Calculate the hardness coefficient Its expression is: in, This indicates the projected area of ​​the grab bucket. This indicates the closing speed of the grab drive mechanism; The vibration sensor collects high-frequency energy during the switching phase. Calculate hardness grade Its expression is: in, The compensation factor represents the torque threshold of the grab drive mechanism during the switching phase. The compensation factor for the grab speed parameter during the grab switching phase. , , Indicates the weighting coefficient; According to the hardness coefficient and hardness rating The coal hardness is statistically graded, and a hardware grading correspondence table is constructed as real-time sensing data for coal hardness.

[0006] As a preferred embodiment of the present invention, the real-time sensing data is input into a hardness-control parameter mapping table, and the corresponding target operating parameters are queried and output, including: The real-time sensing data is input into the hardness-control parameter mapping table, which presets multiple sets of basic operating parameters and corresponding dynamic weight coefficients for different hardness ranges. The perceived data is processed using a predefined fuzzy rule base to infer recommended values ​​for one or more target operation parameters.

[0007] As a preferred embodiment of the present invention, a fuzzy rule base is used to map the nonlinear relationship between the target operating parameters and hardness. The target operating parameters include at least the grab bucket cutting speed, the closed bucket motor torque threshold, or the lifting speed, including: Based on real-time sensing data and historical optimal operation parameters, the weights of fuzzy rules are dynamically adjusted using reinforcement learning algorithms; The construction of the fuzzy rule base is specifically as follows: coal hardness is defined as three fuzzy linguistic variables: soft, medium, and hard, and a corresponding membership function is set for each variable; The grab cutting speed, the closing motor torque threshold, and the lifting speed are defined as three fuzzy linguistic variables: low, medium, and high, respectively.

[0008] As a preferred embodiment of the present invention, the fuzzy rule base specifically comprises: The fuzzy rule base adopts IF-THEN fuzzy rules, with the premise part consisting of fuzzy linguistic variables of coal hardness and the conclusion part consisting of fuzzy linguistic variables of the target operation parameters. Through the fuzzy inference, precise real-time sensing data is transformed into fuzzy hardness levels, the corresponding fuzzy rules are activated, and the inference results are defuzzified into precise target operation parameter values. Based on long-term operating efficiency and energy consumption data, the parameters of the membership function are optimized, and the fuzzy rule base is updated in real time.

[0009] As a preferred embodiment of the present invention, the target operating parameters are dynamically compensated in real time by incorporating ambient temperature and moisture content, including: Based on the ambient temperature, compensation is made for the grab's cutting speed and lifting speed: When the ambient temperature is lower than the preset low temperature threshold, the system determines that the coal viscosity increases and it is easy to freeze and adhere. At this time, the compensation factor of the speed parameter is increased. When the ambient temperature is higher than the preset high temperature threshold, the system takes into account the possible decrease in hydraulic system efficiency and equipment heat load, and at this time reduces the compensation factor of speed parameter. Based on the moisture content, the torque threshold of the closed-bucket motor is compensated: When the moisture content is higher than the preset high humidity threshold, the compensation factor of the torque threshold is increased accordingly to cope with the enhanced coal adhesion caused by high humidity, while the upper limit of the cutting depth of the coal hardness grade is reduced by 20%.

[0010] As a preferred embodiment of the present invention, the dynamic compensation mechanism is implemented by expanding the fuzzy rule base, including: The ambient temperature and moisture content are also used as input variables of the fuzzy inference system, together with the coal hardness, to form the premise of the multidimensional fuzzy rule. A dynamic compensator is established, which receives the ambient temperature and moisture content as input. The dynamic compensator calculates one or more compensation factors for correcting the target operating parameters based on the ambient temperature and moisture content. The target operation parameters are multiplied by the corresponding compensation factors to output the final operation parameters after real-time correction. The target operation parameters that have been comprehensively compensated are directly output through a single inference process.

[0011] As a preferred embodiment of the present invention, based on the dynamic compensation mechanism, the grab unloader is controlled to perform subsequent grabbing operations with the target operating parameters through the conclusion part of the fuzzy rule base, including: The target operation parameters or the final operation parameters after dynamic compensation are converted into corresponding control commands; The control commands are sent to the grab bucket drive mechanism, the bucket closing mechanism, and the lifting mechanism of the ship unloader, controlling each mechanism to operate with the parameters to complete a single grabbing operation cycle.

[0012] As a preferred embodiment of the present invention, monitoring the unloading efficiency per unit time and using the unloading efficiency as a feedback signal includes: The historical best operating data or theoretical calculation value is used as the benchmark unloading efficiency; the deviation between the real-time monitored unloading efficiency per unit time and the benchmark unloading efficiency is calculated. When the deviation exceeds the preset threshold, the operating parameters mapped to the corresponding hardness range in the hardness-control parameter mapping table are adjusted in the same direction according to the direction and magnitude of the deviation.

[0013] An expert rule base is introduced, and the unloading efficiency feedback signal is input into the expert rule base. The expert rule base predefines parameter correction strategies for different efficiency deviation phenomena and their corresponding parameters. The system corrects the mapping table by matching the strategies in the expert rule base according to the current efficiency deviation. The strategy is as follows: if efficiency decreases and the bucket full rate is low, increase the torque threshold of the closed bucket motor; if efficiency decreases and the average load current of the equipment is too high, appropriately reduce the cutting speed.

[0014] As a preferred embodiment of the present invention, the unloading efficiency is used as a reward signal, and a reinforcement learning algorithm is employed to adaptively optimize the hardness-control parameter mapping table, including: The selection of different hardness states and combinations of operating parameters is defined as the action of the agent, and the improvement of unloading efficiency is defined as a positive reward. The strategy function of the mapping table is continuously updated using a reinforcement learning algorithm, so that the system dynamically converges to the optimal mapping relationship that yields the highest unloading efficiency or the lowest overall energy consumption.

[0015] Compared with the prior art, the present invention has the following advantages: This invention dynamically adjusts key operating parameters of the unloader, such as the grab bucket cutting speed, bucket closing torque, and lifting speed, by sensing the hardness of the coal in real time. It also sets dynamic thresholds for these key operating parameters, fundamentally avoiding brute-force operation in hard coal conditions. This effectively reduces the risk of grab bucket deformation, wire rope wear, and damage to the transmission mechanism. Furthermore, it introduces a dynamic compensation mechanism for ambient temperature and moisture content, enabling it to intelligently cope with complex operating conditions such as low-temperature bucket sticking and enhanced adhesion in high humidity. By automatically adjusting parameters, it achieves proactive protection of the equipment, significantly extending the service life of core components and reducing maintenance costs and failure rates.

[0016] By employing fuzzy reasoning and multi-sensor information fusion technology, the system can not only handle the core variable of coal hardness, but also comprehensively consider environmental factors such as temperature and humidity, achieving multi-dimensional and nonlinear intelligent decision-making. It simulates and surpasses the judgment logic of human experts. By integrating expert rules and reinforcement learning, the system can continuously learn and evolve from operational data, automatically discover the optimal control strategy and adapt to changes in operating conditions. This achieves a fundamental transformation from static automated execution to dynamic autonomous optimization, constructing an adaptive closed-loop control system. It effectively overcomes the problems of low efficiency and equipment wear caused by changes in coal quality in the traditional fixed-parameter unloading mode, achieving a significant improvement in unloading efficiency, an extension of equipment life, and a reduction in overall energy consumption, thus achieving the goal of intelligent operation. Attached Figure Description

[0017] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0018] Figure 1 The flowchart illustrates the adaptive unloading method based on coal hardness provided in this embodiment of the invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, this invention provides an adaptive unloading method based on coal hardness, comprising the following steps: The closing speed of the grab bucket drive mechanism of the ship unloader is detected in real time, and the grab bucket pressure value is collected through the hydraulic cylinder pressure sensor. The real-time sensing data characterizing the hardness of coal is calculated based on the grab bucket pressure value. In this embodiment, dynamic thresholds are set for key parameters such as the torque of the closed bucket motor based on the real-time perceived coal hardness, rather than fixed upper limits. This avoids "brute force" operation under hard coal conditions, fundamentally reduces the deformation rate of the grab bucket and tooth tips, as well as the overload fatigue damage of the wire rope and transmission mechanism, significantly extends the service life of core components, and reduces maintenance costs and safety risks.

[0021] The real-time sensing data is input into the hardness-control parameter mapping table, and the corresponding target operation parameters are queried and output. The target operation parameters include at least the grab bucket cutting speed, the closed bucket motor torque threshold, or the lifting speed. By introducing variables such as ambient temperature and moisture content, the target operating parameters are dynamically compensated in real time to correct them, and the grab unloader is controlled to perform subsequent grabbing operations with the target operating parameters. In this embodiment, not only does it respond to the core variable of coal hardness, but it also introduces variables such as ambient temperature and moisture content for dynamic compensation, making the control strategy more refined and more in line with the complex and ever-changing actual working conditions.

[0022] The unloading efficiency per unit time is monitored, and the unloading efficiency is used as a feedback signal to adaptively correct the hardness-control parameter mapping table.

[0023] In this embodiment, by sensing the hardness of the coal in real time and dynamically adjusting the operating parameters, "coal-specific unloading" is achieved. Under soft coal conditions, more aggressive parameters, such as higher cutting speed, are automatically adopted to increase the workload and speed of a single cycle. Under hard coal conditions, robust parameters are intelligently matched to avoid overload shutdown. Thus, the equipment can approach its optimal capacity under different coal qualities, effectively solving the problems of throughput reduction of more than 20% under soft coal conditions and frequent shutdowns under hard coal conditions, and significantly improving overall operating efficiency.

[0024] In this embodiment, the mapping relationship is continuously corrected through feedback on unloading efficiency, enabling the system to continuously optimize itself and gradually accumulate the optimal operating strategy. This completely changes the traditional model that relies on the experience of experienced operators, realizes the automatic evolution and accumulation of control strategies, and achieves a qualitative leap in the level of intelligence.

[0025] The closing speed of the grab bucket drive mechanism of the ship unloader is detected in real time. The grab bucket pressure value is collected through a hydraulic cylinder pressure sensor. Real-time sensing data characterizing coal hardness is calculated based on the grab bucket pressure value, including: The closing speed of the grab bucket drive mechanism of the ship unloader is monitored in real time in two stages: the grab bucket closing stage and the switching stage. In this embodiment, by combining the quasi-static pressure signal of the closing phase and the vibration signal of the switching phase, the limitations of a single signal source are overcome, making the hardness perception more comprehensive and accurate. The phased processing can effectively avoid abnormal interference that may occur in a single phase, thus improving the robustness of the system. During the grab closing phase, the peak value of the hydraulic cylinder pressure sensor is collected. Calculate the hardness coefficient Its expression is: in, This indicates the projected area of ​​the grab bucket. This indicates the closing speed of the grab drive mechanism; The vibration sensor collects high-frequency energy during the switching phase. Calculate hardness grade Its expression is: in, The compensation factor represents the torque threshold of the grab drive mechanism during the switching phase. The compensation factor for the grab speed parameter during the grab switching phase. , , Indicates the weighting coefficient; According to the hardness coefficient and hardness rating The coal hardness is statistically graded, and a hardware grading correspondence table is constructed as real-time sensing data for coal hardness.

[0026] In this embodiment, based on the calculated hardness coefficient and hardness rating Data fusion is performed to classify and statistically grade the coal hardness; specifically, a pre-set fuzzy logic mapping table is used to classify ( , The input combination is mapped to a comprehensive hardness level identifier, such as H1, H2, H3..., and a hardware classification correspondence table is constructed based on this. This table serves as the real-time sensing data that ultimately characterizes the hardness of the coal and is used for subsequent mapping and querying of control parameters. This overcomes the limitations of a single signal source and makes hardness sensing more comprehensive and accurate.

[0027] The real-time sensing data is input into the hardness-control parameter mapping table, and the corresponding target operation parameters are queried and output, including: The real-time sensing data is input into the hardness-control parameter mapping table, which presets multiple sets of basic operating parameters and corresponding dynamic weight coefficients for different hardness ranges. The perceived data is processed using a predefined fuzzy rule base to infer recommended values ​​for one or more target operation parameters.

[0028] In this embodiment, the real-time sensing data is input into the hardness-control parameter mapping table, and the corresponding target operation parameters are queried and output. Specifically, this includes the following steps: S1. Dynamic Weight Mapping Table Query: Input the real-time sensing data, i.e., the comprehensive hardness grade identifier, into the hardness-control parameter mapping table for querying. The structure of the mapping table is as follows: For each discrete hardness grade interval, such as H1, H2, H3..., a set of basic operating parameters are preset, including the basic cutting speed. Basic closed bucket torque threshold Basic improvement speed and a set of dynamic weight coefficients associated with it. , and ; S2. Fuzzy Inference Parameter Optimization: The real-time sensing data and its derived working condition context information are input into a predefined fuzzy rule base for processing. The inference mechanism of this fuzzy rule base is as follows: Input variables: The core input is the comprehensive hardness level identifier, and auxiliary input variables such as "current full load rate of grab bucket" and "historical average working cycle" are introduced. Fuzzy reasoning: Using predefined fuzzy rules, for example: if the hardness is H2 and the full load rate is low, the weight of THEN lifting speed is slightly increased; if the hardness is H3 and the operation cycle is relatively long, the weight of THEN cutting speed is slightly decreased, and fuzzy logic reasoning is performed on these inputs. Output: The inference output is the optimized adjustment amount of the aforementioned dynamic weight coefficients; S3. Weighted Fusion Output of Final Parameters: The basic job parameters obtained from the query are weighted and fused with the dynamic weight coefficients optimized by fuzzy inference to calculate the final target job parameters. The calculation formula is as follows: Final target operation parameters = basic operation parameters × optimized dynamic weight coefficient In this embodiment, by using a structure of basic parameters + dynamic weights, fuzzy inference is introduced to optimize the weights in real time, enabling the output parameters to respond sensitively to subtle changes in operating conditions beyond hardness, thus achieving higher-dimensional adaptive control.

[0029] In this embodiment, the basic mapping table ensures the stability and security of the control strategy at the macro level, while fuzzy inference gives the system the ability to make flexible adjustments at the micro level. This combination ensures that the system will not oscillate due to oversensitivity, nor will it miss optimization opportunities due to excessive rigidity.

[0030] A fuzzy rule base is used to map the nonlinear relationship between the target operating parameters and hardness. The target operating parameters include at least the grab bucket cutting speed, the closed bucket motor torque threshold, or the lifting speed, including: Based on real-time sensing data and historical optimal operation parameters, the weights of fuzzy rules are dynamically adjusted using reinforcement learning algorithms; The construction of the fuzzy rule base is specifically as follows: coal hardness is defined as three fuzzy linguistic variables: soft, medium, and hard, and a corresponding membership function is set for each variable; The grab cutting speed, the closing motor torque threshold, and the lifting speed are defined as three fuzzy linguistic variables: low, medium, and high, respectively.

[0031] The fuzzy rule base is specifically as follows: The fuzzy rule base adopts IF-THEN fuzzy rules, with the premise part consisting of fuzzy linguistic variables of coal hardness and the conclusion part consisting of fuzzy linguistic variables of the target operation parameters. Through the fuzzy inference, precise real-time sensing data is transformed into fuzzy hardness levels, the corresponding fuzzy rules are activated, and the inference results are defuzzified into precise target operation parameter values. Based on long-term operating efficiency and energy consumption data, the parameters of the membership function are optimized, and the fuzzy rule base is updated in real time.

[0032] In this embodiment, the real-time collected precise sensing data is transformed into a fuzzy set of corresponding input variables. Based on the fuzzified input, the matching fuzzy rules in the rule base are activated. The fuzzy conclusions inferred from multiple rules are aggregated and transformed into precise target operation parameter values ​​that can be directly executed through defuzzification algorithms such as the centroid method.

[0033] In this embodiment, based on long-term statistical data on work efficiency and energy consumption, and with the goal of maximizing overall efficiency, a genetic algorithm is used to automatically correct the membership function parameters of the input / output variables, making the fuzzy division more consistent with actual working conditions.

[0034] By incorporating ambient temperature and moisture content, a dynamic compensation mechanism is used to correct the target operating parameters in real time, including: Based on the ambient temperature, compensation is made for the grab's cutting speed and lifting speed: When the ambient temperature is lower than the preset low temperature threshold, the system determines that the coal viscosity increases and it is easy to freeze and adhere. At this time, the compensation factor of the speed parameter is increased. When the ambient temperature is higher than the preset high temperature threshold, the system takes into account the possible decrease in hydraulic system efficiency and equipment heat load, and at this time reduces the compensation factor of speed parameter. Based on the moisture content, the torque threshold of the closed-bucket motor is compensated: When the moisture content is higher than the preset high humidity threshold, the compensation factor of the torque threshold is increased accordingly to cope with the enhanced coal adhesion caused by high humidity, while the upper limit of the cutting depth of the coal hardness grade is reduced by 20%.

[0035] In this embodiment, when the ambient temperature is below 0°C, the system determines that the coal is more viscous and prone to freezing and adhesion. At this time, the compensation factor of the speed parameter is increased, such as multiplying the speed reference value by a coefficient greater than 1, in order to overcome the adhesion force with higher kinetic energy and ensure that the material can be smoothly unloaded from the grab bucket.

[0036] In this embodiment, when the ambient temperature is higher than 35°C, the system considers the possible decrease in hydraulic system efficiency and equipment heat load. At this time, the compensation factor of the speed parameter is reduced, such as multiplying the speed reference value by a coefficient less than 1, and the speed under hardness grade conditions is reduced by 10% to achieve equipment protection and energy-saving operation.

[0037] In this embodiment, when the moisture content is higher than 15%, the system determines that the coal's stickiness and adhesion are significantly enhanced, making it difficult for the grab bucket to fill and unload. At this time, the compensation factor for the torque threshold of the closing bucket motor is increased to give the grab bucket a greater closing force to overcome resistance and ensure the full bucket rate. At the same time, in order to prevent the grab bucket from being "sucked" or overloaded due to excessive cutting in high-moisture sticky materials, the upper limit of the maximum cutting depth allowed for the corresponding coal hardness grade is dynamically reduced by 20%. By limiting the cutting amount, the increase in torque is balanced to achieve safe and efficient grabbing.

[0038] In this embodiment, the single control mode that relies solely on coal hardness is transcended. By introducing two key operating variables, ambient temperature and moisture content, multi-factor collaborative decision-making is achieved. This expands the control strategy from "one-dimensional" to "three-dimensional," enabling more precise responses to actual changes in the physical properties of coal. This dynamic balancing mechanism ensures that while pursuing high efficiency, the system always prioritizes the safe operation of the equipment.

[0039] In this embodiment, an automated solution is provided to address common pain points in field operations such as "bucket sticking in winter" and "difficulty digging in the rainy season". This solution can significantly reduce manual intervention, stabilize work efficiency, and reduce the risk of equipment failure due to material characteristics.

[0040] The dynamic compensation mechanism is implemented by expanding the fuzzy rule base, including: The ambient temperature and moisture content are also used as input variables of the fuzzy inference system, together with the coal hardness, to form the premise of the multidimensional fuzzy rule. A dynamic compensator is established, which receives the ambient temperature and moisture content as input. In this embodiment, instead of simply multiplying and superimposing the compensation factors of each variable mechanically, the system uses a fuzzy rule base to perform deep coupling and intelligent decision-making at the logical level. The system can understand that the required operating strategies are fundamentally different "in hard and wet materials" and "in hard and dry materials", thereby outputting a truly optimal and integrated control command, rather than a simple splicing of several independent compensation results.

[0041] The dynamic compensator calculates one or more compensation factors for correcting the target operating parameters based on the ambient temperature and moisture content. The target operation parameters are multiplied by the corresponding compensation factors to output the final operation parameters after real-time correction. The target operation parameters that have been comprehensively compensated are directly output through a single inference process.

[0042] In this embodiment, the controller and dynamic compensator are integrated into one, and all decisions are made through a unified fuzzy inference engine. This avoids the system complexity caused by establishing multiple independent compensation models, reduces the interfaces and data transmission between modules, improves the real-time performance of control response, reduces the difficulty of system maintenance and potential failure points, and enhances overall reliability.

[0043] Based on the dynamic compensation mechanism, the grab unloader is controlled to perform subsequent grabbing operations with the target operating parameters through the conclusion part of the fuzzy rule base, including: The target operation parameters or the final operation parameters after dynamic compensation are converted into corresponding control commands; In this embodiment, the final operating parameters obtained after fuzzy reasoning and dynamic compensation, including precise grab bucket cutting speed setting value, bucket closing motor torque threshold setting value, and lifting speed setting value, are converted into corresponding analog control commands by the dedicated motion controller of the ship unloader. The control commands are then sent in real time and synchronously to the drivers or execution units of the grab bucket drive mechanism, bucket closing mechanism, and lifting mechanism of the ship unloader via an industrial fieldbus.

[0044] The control commands are sent to the grab bucket drive mechanism, the bucket closing mechanism, and the lifting mechanism of the ship unloader, controlling each mechanism to operate with the parameters to complete a single grabbing operation cycle.

[0045] In this embodiment, an adaptive control method is adopted to convert the optimal decision made by the upper-level intelligent algorithm into the precise actions of the lower-level equipment without distortion and with high fidelity. This ensures that all the results of the aforementioned intelligent sensing and optimization calculation based on coal hardness, temperature and humidity can be fully and efficiently reflected in the actual unloading operation, forming a complete intelligent closed loop from sensing to execution.

[0046] In this embodiment, it is ensured that multiple mechanisms such as grab bucket drive, bucket closing, and lifting can receive the same set of parameters calculated within the same control cycle. This avoids incoordination of actions, mechanism interference, or efficiency loss caused by asynchronous parameter updates, making the entire grabbing cycle smooth and significantly improving the smoothness of operation and equipment stability.

[0047] Monitoring the unloading efficiency per unit time and using the unloading efficiency as a feedback signal includes: The historical best operating data or theoretical calculation value is used as the benchmark unloading efficiency; the deviation between the real-time monitored unloading efficiency per unit time and the benchmark unloading efficiency is calculated. When the deviation exceeds the preset threshold, the operating parameters mapped to the corresponding hardness range in the hardness-control parameter mapping table are adjusted in the same direction according to the direction and magnitude of the deviation.

[0048] An expert rule base is introduced, and the unloading efficiency feedback signal is input into the expert rule base. The expert rule base predefines parameter correction strategies for different efficiency deviation phenomena and their corresponding parameters. The system corrects the mapping table by matching the strategies in the expert rule base according to the current efficiency deviation. The strategy is as follows: if efficiency decreases and the bucket full rate is low, increase the torque threshold of the closed bucket motor; if efficiency decreases and the average load current of the equipment is too high, appropriately reduce the cutting speed.

[0049] In this embodiment, through intelligent closed-loop optimization, the system can dynamically adjust itself to the optimal operating point. It can not only find the optimal solution under the current working conditions, but also automatically track and maintain this optimal solution as long-term factors such as equipment wear, seasonal changes or changes in coal sources occur. This ensures that the ship unloader can maintain efficient, low-consumption and safe operation throughout its entire life cycle, greatly improving the long-term comprehensive efficiency and value of the equipment.

[0050] Using the unloading efficiency as a reward signal, a reinforcement learning algorithm is used to adaptively optimize the stiffness-control parameter mapping table, including: The selection of different hardness states and combinations of operating parameters is defined as the action of the agent, and the improvement of unloading efficiency is defined as a positive reward. The strategy function of the mapping table is continuously updated using a reinforcement learning algorithm, so that the system dynamically converges to the optimal mapping relationship that yields the highest unloading efficiency or the lowest overall energy consumption.

[0051] In this embodiment, the optimal strategy after learning convergence is directly reflected in the dynamic updating and overlay of the hardness-control parameter mapping table, so that the entire system can dynamically converge and automatically maintain the global optimal mapping relationship with the highest unloading efficiency or the lowest overall operating cost. This guides the system to pursue high efficiency while taking into account equipment protection and energy saving, thereby maximizing the overall performance of the equipment throughout its entire life cycle.

[0052] This invention dynamically adjusts key operating parameters of the unloader, such as the grab bucket cutting speed, bucket closing torque, and lifting speed, by sensing the hardness of the coal in real time. It also sets dynamic thresholds for these key operating parameters, fundamentally avoiding brute-force operation in hard coal conditions. This effectively reduces the risk of grab bucket deformation, wire rope wear, and damage to the transmission mechanism. Furthermore, it introduces a dynamic compensation mechanism for ambient temperature and moisture content, enabling it to intelligently cope with complex operating conditions such as low-temperature bucket sticking and enhanced adhesion in high humidity. By automatically adjusting parameters, it achieves proactive protection of the equipment, significantly extending the service life of core components and reducing maintenance costs and failure rates.

[0053] By employing fuzzy reasoning and multi-sensor information fusion technology, the system can not only handle the core variable of coal hardness, but also comprehensively consider environmental factors such as temperature and humidity, achieving multi-dimensional and nonlinear intelligent decision-making. It simulates and surpasses the judgment logic of human experts. By integrating expert rules and reinforcement learning, the system can continuously learn and evolve from operational data, automatically discover the optimal control strategy and adapt to changes in operating conditions. This achieves a fundamental transformation from static automated execution to dynamic autonomous optimization, constructing an adaptive closed-loop control system. It effectively overcomes the problems of low efficiency and equipment wear caused by changes in coal quality in the traditional fixed-parameter unloading mode, achieving a significant improvement in unloading efficiency, an extension of equipment life, and a reduction in overall energy consumption, thus achieving the goal of intelligent operation.

[0054] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. An adaptive unloading method based on coal hardness, characterized in that, Includes the following steps: The closing speed of the grab bucket drive mechanism of the ship unloader is detected in real time, and the grab bucket pressure value is collected through the hydraulic cylinder pressure sensor. The real-time sensing data characterizing the hardness of coal is calculated based on the grab bucket pressure value. The real-time sensing data is input into the hardness-control parameter mapping table, and the corresponding target operation parameters are queried and output. The target operation parameters include at least the grab bucket cutting speed, the closed bucket motor torque threshold, or the lifting speed. By introducing variables such as ambient temperature and moisture content, the target operation parameters are dynamically compensated in real time to correct them, and the grab unloader is controlled to perform subsequent grabbing operations with the target operation parameters. The unloading efficiency per unit time is monitored, and the unloading efficiency is used as a feedback signal to adaptively correct the hardness-control parameter mapping table.

2. The adaptive unloading method based on coal hardness according to claim 1, characterized in that, The closing speed of the grab bucket drive mechanism of the ship unloader is detected in real time. The grab bucket pressure value is collected through a hydraulic cylinder pressure sensor. Real-time sensing data characterizing coal hardness is calculated based on the grab bucket pressure value, including: The closing speed of the grab bucket drive mechanism of the ship unloader is monitored in real time in two stages: the grab bucket closing stage and the switching stage. During the grab closing phase, the peak value of the hydraulic cylinder pressure sensor is collected. Calculate the hardness coefficient Its expression is: in, This indicates the projected area of ​​the grab bucket. This indicates the closing speed of the grab drive mechanism; The vibration sensor collects high-frequency energy during the switching phase. Calculate hardness grade Its expression is: in, The compensation factor represents the torque threshold of the grab drive mechanism during the switching phase. The compensation factor for the grab speed parameter during the grab switching phase. , , Indicates the weighting coefficient; According to the hardness coefficient and hardness rating The coal hardness is statistically graded, and a hardware grading correspondence table is constructed as real-time sensing data for coal hardness.

3. The adaptive unloading method based on coal hardness according to claim 2, characterized in that, The real-time sensing data is input into the hardness-control parameter mapping table, and the corresponding target operation parameters are queried and output, including: The real-time sensing data is input into the hardness-control parameter mapping table, which presets multiple sets of basic operating parameters and corresponding dynamic weight coefficients for different hardness ranges. The perceived data is processed using a predefined fuzzy rule base to infer recommended values ​​for one or more target operation parameters.

4. The adaptive unloading method based on coal hardness according to claim 3, characterized in that, A fuzzy rule base is used to map the nonlinear relationship between the target operating parameters and hardness. The target operating parameters include at least the grab bucket cutting speed, the closed bucket motor torque threshold, or the lifting speed, including: Based on real-time sensing data and historical optimal operation parameters, the weights of fuzzy rules are dynamically adjusted using reinforcement learning algorithms; The construction of the fuzzy rule base is specifically as follows: coal hardness is defined as three fuzzy linguistic variables: soft, medium, and hard, and a corresponding membership function is set for each variable; The grab cutting speed, the closing motor torque threshold, and the lifting speed are defined as three fuzzy linguistic variables: low, medium, and high, respectively.

5. The adaptive unloading method based on coal hardness according to claim 3, characterized in that, The fuzzy rule base is specifically as follows: The fuzzy rule base adopts IF-THEN fuzzy rules, with the premise part consisting of fuzzy linguistic variables of coal hardness and the conclusion part consisting of fuzzy linguistic variables of the target operation parameters. Through the fuzzy inference, precise real-time sensing data is transformed into fuzzy hardness levels, the corresponding fuzzy rules are activated, and the inference results are defuzzified into precise target operation parameter values. Based on long-term operating efficiency and energy consumption data, the parameters of the membership function are optimized, and the fuzzy rule base is updated in real time.

6. The adaptive unloading method based on coal hardness according to claim 5, characterized in that, By incorporating ambient temperature and moisture content, a dynamic compensation mechanism is used to correct the target operating parameters in real time, including: Based on the ambient temperature, compensation is made for the grab's cutting speed and lifting speed: When the ambient temperature is lower than the preset low temperature threshold, the system determines that the coal viscosity increases and it is easy to freeze and adhere. At this time, the compensation factor of the speed parameter is increased. When the ambient temperature is higher than the preset high temperature threshold, the system takes into account the possible decrease in hydraulic system efficiency and equipment heat load, and at this time reduces the compensation factor of speed parameter. Based on the moisture content, the torque threshold of the closed-bucket motor is compensated: When the moisture content is higher than the preset high humidity threshold, the compensation factor of the torque threshold is increased accordingly to cope with the enhanced coal adhesion caused by high humidity, while the upper limit of the cutting depth of the coal hardness grade is reduced by 20%.

7. The adaptive unloading method based on coal hardness according to claim 6, characterized in that, The dynamic compensation mechanism is implemented by expanding the fuzzy rule base, including: The ambient temperature and moisture content are also used as input variables of the fuzzy inference system, together with the coal hardness, to form the premise of the multidimensional fuzzy rule. A dynamic compensator is established, which receives the ambient temperature and moisture content as input. The dynamic compensator calculates one or more compensation factors for correcting the target operating parameters based on the ambient temperature and moisture content. The target operation parameters are multiplied by the corresponding compensation factors to output the final operation parameters after real-time correction. The target operation parameters that have been comprehensively compensated are directly output through a single inference process.

8. The adaptive unloading method based on coal hardness according to claim 7, characterized in that, Based on the dynamic compensation mechanism, the grab unloader is controlled to perform subsequent grabbing operations with the target operating parameters through the conclusion part of the fuzzy rule base, including: The target operation parameters or the final operation parameters after dynamic compensation are converted into corresponding control commands; The control commands are sent to the grab bucket drive mechanism, the bucket closing mechanism, and the lifting mechanism of the ship unloader, controlling each mechanism to operate with the parameters to complete a single grabbing operation cycle.

9. The adaptive unloading method based on coal hardness according to claim 8, characterized in that, Monitoring the unloading efficiency per unit time and using the unloading efficiency as a feedback signal includes: The historical best operating data or theoretical calculation value is used as the benchmark unloading efficiency; the deviation between the real-time monitored unloading efficiency per unit time and the benchmark unloading efficiency is calculated. When the deviation exceeds the preset threshold, the working parameters mapped to the corresponding hardness range in the hardness-control parameter mapping table are adjusted in the same direction according to the direction and magnitude of the deviation. An expert rule base is introduced, and the unloading efficiency feedback signal is input into the expert rule base. The expert rule base predefines parameter correction strategies for different efficiency deviation phenomena and their corresponding parameters. The system corrects the mapping table by matching the strategies in the expert rule base according to the current efficiency deviation. The strategy is as follows: if efficiency decreases and the bucket full rate is low, increase the torque threshold of the closed bucket motor; if efficiency decreases and the average load current of the equipment is too high, appropriately reduce the cutting speed.

10. The adaptive unloading method based on coal hardness according to claim 9, characterized in that, Using the unloading efficiency as a reward signal, a reinforcement learning algorithm is used to adaptively optimize the stiffness-control parameter mapping table, including: The selection of different hardness states and combinations of operating parameters is defined as the action of the agent, and the improvement of unloading efficiency is defined as a positive reward. The strategy function of the mapping table is continuously updated using a reinforcement learning algorithm, so that the system dynamically converges to the optimal mapping relationship that yields the highest unloading efficiency or the lowest overall energy consumption.

Images