A coke production four-car full-automatic dynamic scheduling control system

Abstract

The present application relates to the technical fields of industrial automation control, and discloses a coke production four-car full-automatic dynamic scheduling control system for coordinating and controlling the automatic operation of the four-car composed of SCP integrated machine, coke blocking car, smoke guide car and coke tank car, comprising a vehicle state sensing module, a cooperative state prediction module, a dynamic scheduling decision module and a scheduling instruction distribution and execution module.In the present application, the real-time collection and standardized processing of the four-car operation data are realized through the vehicle state sensing module, providing comprehensive and accurate data support for scheduling decision, based on the real-time data, the cooperative state prediction module can predict the vehicle readiness time difference and quantify the cooperative stress index, so that the system changes from passive response to active prevention of congestion risk, effectively reducing the invalid waiting of vehicles, when the risk exceeds the threshold, the dynamic scheduling decision module can quickly generate multiple candidate schemes and select the optimal solution through a multi-objective utility function, realizing the real-time optimization of the scheduling strategy.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, and in particular to a fully automatic dynamic scheduling and control system for four coke production vehicles. Background Technology

[0002] In the core operation process of coke production, the coordinated operation of the coke pusher (SCP integrated machine), coke quencher, flue gas guide car, and coke tank car (collectively referred to as the four major cars) directly determines the coke oven production cycle and coke quality. Their operation must strictly follow the continuous process of pushing coke, quenching coke, guiding flue gas, and receiving coke. Currently, the mainstream scheduling control system of the four major cars in the industry mostly adopts the static scheduling mode based on the preset production plan. This mode formulates operation instructions with fixed process sequence as the core. Under ideal conditions of no equipment failure and stable operating conditions, it can ensure the orderly progress of the basic production process and is currently the most widely used basic scheduling scheme.

[0003] The core drawback of the existing static scheduling model lies in the contradiction between its fixed-plan-centric logic and the dynamic nature of coke production. This leads to several problems: The system lacks real-time, comprehensive awareness of the operating status of the four main coke machines, only acquiring discrete information such as vehicle start-up / stop and furnace alignment. It cannot accurately grasp dynamic data such as vehicle position deviations, action delays, and potential equipment failures, resulting in unreliable data support for scheduling decisions. Furthermore, the system lacks the ability to predict collaborative states, failing to anticipate the time difference between the readiness of preceding and following vehicles or quantify path congestion risks. When a vehicle experiences delays in coke pushing or malfunctions, its coordinating vehicles can only passively wait or stop, disrupting the production cycle. The fixed scheduling strategy cannot be flexibly adjusted according to changing operating conditions. After an anomaly occurs, manual intervention is required to re-formulate the plan, resulting in delayed response and susceptibility to misoperation due to information asymmetry. Ultimately, this leads to a significant decrease in equipment utilization, reduced production efficiency, and even potential safety accidents due to vehicle coordination errors. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing four-vehicle dispatching control system for coke production suffers from insufficient real-time perception, poor collaborative prediction capability, rigid dispatching strategy, and lack of multi-objective optimization. To address this, we propose a fully automatic dynamic dispatching control system for the four-vehicle coke production.

[0005] To achieve the above objectives, this application adopts the following technical solution: a fully automatic dynamic scheduling and control system for four major coke production vehicles, used to coordinate and control the automated operation of the four major vehicles consisting of the SCP integrated machine, coke quenching car, flue gas guide car, and coke tank car, including: The vehicle status perception module, based on the encoders, photoelectric switches and travel limit sensors on the four main vehicle bodies and the track, collects real-time operating data on the vehicle's absolute position, alignment with furnace number, movement trolley limit status and travel inverter status. The collaborative status prediction module is used to receive data from the vehicle status perception module, integrate historical operating data and real-time operating conditions, predict the preparation time of preceding and subsequent vehicles on key process paths through time series analysis, and calculate the collaborative pressure index that reflects the system's collaborative efficiency and congestion risk. The dynamic scheduling decision module continuously monitors the collaborative pressure index. When it exceeds the safety threshold for dynamic adjustment of production cycle time, the strategy engine is activated. Based on the rule base, candidate solutions such as insertion waiting, step skipping, and path replanning are generated. Through multi-objective utility function evaluation and scoring that integrates time, energy consumption, and system stability, the globally optimal scheduling instruction with the highest comprehensive utility value is selected. The scheduling instruction distribution and execution module is used to parse and encapsulate the optimal scheduling instructions into a standardized instruction set that conforms to the controller protocol, and distribute them concurrently via the industrial Ethernet bus. It monitors the execution status and confirms the agreement through a handshake protocol to ensure that the relevant vehicles reach a consensus before starting collaborative operations.

[0006] Preferably, the cooperative state prediction module specifically includes: The time prediction unit is used to predict the completion time of the preceding vehicle actions and the preparation time for the subsequent vehicle to be in place for the coordinated path from the completion of coke pushing to the coke tank car being in place and receiving coke, and from the dust removal of the smoke guide car to the closing of the coke blocking car. The risk field strength calculation unit, coupled to the time prediction unit, is used to receive the predicted time difference of each cooperative path, and to calculate the system-level cooperative pressure index by combining the process criticality weight of the path, the ideal time window, and the physical blockage state.

[0007] Preferably, the collaborative path is based on the coke production process and is a predefined sequence of operations in the system that is completed in relay by two or more vehicles in a specific order.

[0008] Preferably, the dynamic scheduling decision module specifically includes: The strategy scheme generation unit is used to dynamically combine basic scheduling strategies based on the rule base to generate a non-empty and mutually exclusive set of candidate scheduling schemes when the collaborative pressure index exceeds the limit. The multi-objective evaluation unit, coupled to the strategy scheme generation unit, is used to perform multi-dimensional performance simulation on each candidate scheduling scheme in the set and calculate its comprehensive utility score for total order ranking.

[0009] Preferably, the collaborative pressure index Calculated using the following formula: ,in The total number of critical collaborative paths defined in the system; For the first The inherent risk weight of each collaborative path is pre-set based on the degree of impact of path blockage on overall production. For the first Collaborative paths in time The predicted time difference is the estimated time of readiness of the subsequent vehicle minus the estimated time of readiness of the preceding vehicle. For the first The ideal time difference between the collaborative paths is the optimal time interval for achieving smooth collaboration; For the first The time difference tolerance standard deviation of each collaborative path is used to define the flexible range of time matching; For the first The feasibility impulse function of the cooperative path, when the system state vector When the path is indicated to have a physical space hard blockage, its function value is a penalty coefficient much greater than 1. Otherwise, it is 1.

[0010] Preferably, the objectives evaluated by the multi-objective utility function include at least a time objective, an energy consumption objective, and a stability objective; wherein the evaluation value of the time objective is based on the predicted total cycle time after the scheme is executed; the evaluation value of the energy consumption objective is based on the estimated total energy consumption of all additional travel and actions of all vehicles caused by the scheme; and the evaluation value of the stability objective is based on the total number of state transitions required by the scheme for each vehicle to deviate from its preset standard operating sequence.

[0011] Preferably, the comprehensive utility value Calculated using the following formula: ,in The first one in the current candidate scheduling scheme set One candidate scheduling scheme; To optimize the number of objectives; To optimize the target sequence number; Candidate scheduling schemes In the The original evaluation values ​​on each optimization objective; and The first one in the current candidate scheduling scheme set is the first one. Minimum and maximum evaluation values ​​for each optimization objective; For the first The optimization objective is for the candidate scheduling scheme Dynamic weights; For the first The optimization objective is for the candidate scheduling scheme Dynamic weights, The summation index is used to iterate through the first to last index in the exponent denominator. The optimization objectives are calculated and the corresponding dynamic weights are summed.

[0012] Preferably, the dynamic weight Obtain it using the following formula: ,in yes The normalized value is . ; Candidate scheduling schemes In the The normalized value on the optimization objective is determined according to the... The same normalization rules are used to determine this; It is a sensitivity parameter greater than zero, used to control how sensitive the weights are to differences in target performance.

[0013] Preferably, it also includes a system monitoring and alarm module, used to monitor in real time the decision cycle of the dynamic scheduling decision module, the instruction response success rate of the scheduling instruction distribution and execution module, and the network communication status; The system monitoring and alarm module establishes a multi-level alarm mechanism. When it detects that the device communication timeout, the collaborative pressure index is running at a high level for a long time, or the scheduling command is rejected multiple times, it sends audible and visual alarm information, SMS or work order alarm information to the administrator according to the preset level.

[0014] Preferably, the specific process of the handshake confirmation protocol in the scheduling instruction distribution and execution module is as follows: after broadcasting a new scheduling instruction, wait for and collect the condition ready confirmation signals returned by all relevant vehicle controllers; only after receiving all necessary confirmation signals will the final execution command be sent to each controller; if any vehicle returns a rejection signal or fails to respond within a timeout period, the scheduling will be canceled and an anomaly will be reported.

[0015] The technical effects and advantages of this invention are as follows: In this invention, the vehicle status perception module enables real-time collection and standardized processing of the four major vehicle operation data, providing comprehensive and accurate data support for scheduling decisions. This avoids the blind decision-making caused by the lag in status perception. Based on this real-time data, the collaborative status prediction module can accurately predict the vehicle readiness time difference and quantify the collaborative pressure index, enabling the system to shift from passive response to proactive prevention of congestion risks and effectively reduce ineffective vehicle waiting. When the risk exceeds the threshold, the dynamic scheduling decision module can quickly generate multiple candidate solutions and select the optimal solution through a multi-objective utility function, achieving real-time optimization of the scheduling strategy. This completely solves the problem of rigidity in traditional scheduling. Finally, a handshake confirmation protocol ensures the reliable execution of scheduling instructions, avoiding the drawbacks of manual intervention. This significantly improves the stability and efficiency of the four major vehicle collaborative operations, reduces production energy consumption and safety risks, and comprehensively enhances the automation and intelligence level of coke production. Attached Figure Description

[0016] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts: Figure 1 This is a schematic diagram of the system connection principle of the present invention. Detailed Implementation

[0017] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0018] First embodiment: Reference Figure 1 As shown, this invention provides a fully automated dynamic scheduling and control system for four major coke production vehicles, used to coordinate and control the automated operation of the SCP integrated machine, coke quenching car, flue gas guide car and coke tank car. The system mainly includes a vehicle status perception module, a collaborative status prediction module, a dynamic scheduling decision module and a scheduling instruction distribution and execution module.

[0019] Among them, the vehicle status perception module is the basic data source of the system. This module collects the vehicle's operating status data in real time through various sensors deployed on the four vehicle bodies and tracks.

[0020] Specifically, the absolute position information of the vehicle is acquired in real time through encoders to ensure positioning accuracy; photoelectric switches are used to detect whether the vehicle has reached a specific position or completed a specific action, such as whether the coke tank car has been aligned with the coke receiving port; travel limit sensors are used to monitor the front and rear limit status of each moving trolley to determine whether it is in a safe position; in addition, data such as the status of the travel inverter are collected to understand the vehicle's running speed, acceleration, and potential faults; the collected raw data undergoes outlier removal and format standardization processing, filtering out outlier data from sensor false alarms, and unifying the data from different sensors into a standard format and unit to ensure that an accurate and consistent real-time data source is provided for upper-level decision-making.

[0021] The collaborative state prediction module receives standardized data processed by the vehicle state perception module. Its core function is to predict the preparation and readiness times of preceding and succeeding vehicles on critical process paths and to calculate the collaborative pressure index. Specifically, the collaborative state prediction module includes a time prediction unit and a risk field strength calculation unit. The time prediction unit targets critical collaborative paths in coke production, such as the time from coke pushing completion to coke tank car positioning and receiving, or the time from dust removal by the flue car to the closing of the coke quenching car. It utilizes time series analysis methods, preferably selecting an ARIMA model or an LSTM neural network, combining historical operating data with real-time operating conditions. Historical operating data can include average operating time, failure rate, etc., while real-time operating conditions... The conditions can include the current furnace temperature, coke type, etc., to predict the completion time of the preceding vehicle's actions and the preparation time for the subsequent vehicle to be in position. For example, predicting the time for the coke pusher to complete pushing the coke and the time for the coke tank car to arrive at the designated position and be ready to receive the coke. The risk field strength calculation unit is coupled to the time prediction unit to receive the predicted time difference of each path. This unit comprehensively considers the process criticality weight of the path, the ideal time window, and the physical blockage status to calculate the system's collaborative pressure index. This index is a comprehensive quantitative indicator that reflects the overall collaborative efficiency of the system and the potential blockage risk. The greater the deviation of the predicted time difference from the ideal time window, or the existence of physical blockage, the higher the collaborative pressure index.

[0022] The dynamic scheduling decision module continuously monitors the collaborative pressure index. When this index exceeds the safety threshold dynamically adjusted according to the production cycle, the module activates its internal strategy engine. Based on a pre-set rule base, the strategy engine instantly generates multiple candidate scheduling schemes, including insertion waiting, step skipping, and path replanning. For example, if coke tank cars are expected to be delayed, the strategy engine may generate a scheme to insert and wait the coke pushing cars, or a path replanning scheme to adjust the coke pushing order and push other furnace numbers first. The dynamic scheduling decision module includes a strategy scheme generation unit and a multi-objective evaluation unit, where the strategy scheme generation unit is responsible for dynamic scheduling. The basic scheduling strategy of state combination generates a non-empty and mutually exclusive set of candidate scheduling schemes. The multi-objective evaluation unit is coupled to the strategy scheme generation unit and uses a multi-objective utility function that comprehensively considers time, energy consumption and system stability to evaluate and quantify all feasible schemes in parallel. The time objective evaluates the predicted total cycle time after the execution of the scheme; the energy consumption objective evaluates the estimated total energy consumption of all vehicles' additional travel and actions caused by the scheme; the stability objective evaluates the total number of state transitions required for each vehicle to deviate from its preset standard operation sequence; finally, the scheme with the highest comprehensive utility value is selected as the globally optimal scheduling instruction.

[0023] The scheduling instruction distribution and execution module is responsible for parsing and encapsulating the optimal scheduling instructions output by the dynamic scheduling decision module into a standardized instruction set conforming to the communication protocols of the four major vehicle controllers through the instruction parsing and encapsulation unit. These instructions are concurrently sent to the corresponding vehicle controllers via the industrial Ethernet bus. This module also monitors the execution status of the instructions through the instruction execution status monitoring unit, and ensures that all relevant vehicles reach a consensus on the new work plan before starting collaborative work through a handshake confirmation protocol unit. Specifically, after broadcasting a new scheduling instruction, it waits for and collects condition-ready confirmation signals from all relevant vehicle controllers. Only after receiving all necessary confirmation signals is the final execution command sent to each controller. If any vehicle returns a rejection signal or fails to respond within a timeout period, the scheduling is canceled and an anomaly is reported.

[0024] Second embodiment: Based on the first embodiment, this embodiment further describes in detail the internal working mechanism of the collaborative state prediction module and the dynamic scheduling decision module.

[0025] The time prediction unit in the collaborative state prediction module, targeting key collaborative paths in coke production, such as the time from coke pushing completion to coke tank car positioning and coke receiving, and the time from flue dust removal by the flue car to coke quenching car closing, predicts the completion time of the preceding vehicle's actions and the preparation time for the subsequent vehicle's positioning. These key collaborative paths are predefined in the system based on the coke production process and are work sequences that must be completed by two or more vehicles in a specific order. For example, after the coke pushing car finishes pushing the coke, it must wait for the coke tank car to be positioned before receiving the coke; after the flue dust removal by the flue car, the coke quenching car can close the coke quenching door. The time prediction unit analyzes the average operating time, downtime, acceleration and deceleration time of each vehicle in historical data, combined with the current real-time position, speed, target furnace number, and other information of the vehicle, and uses a machine learning model for prediction. The preferred machine learning model is a recurrent neural network (RNN) or a long short-term memory network (LSTM).

[0026] The risk field strength calculation unit is coupled to the time prediction unit to receive the prediction time difference of each path; Coordinated pressure index Calculated using the following formula: ,in The total number of critical collaborative paths defined in the system; For the first The inherent risk weight of each collaborative path is pre-set based on the degree of impact of the path blockage on the overall production line. For example, the weight of the coking path will be higher than that of other paths. For the first Collaborative paths in time The predicted time difference is the estimated time of readiness of the subsequent vehicle minus the estimated time of readiness of the preceding vehicle. For the first The ideal time difference between the two collaborative paths is the optimal time interval for achieving smooth collaboration, and is typically 0 or a very small positive value. For the first The time difference tolerance standard deviation of each collaborative path is used to define the flexible range of time matching; For the first The feasibility impulse function of the cooperative path, when the system state vector When the path is indicated to have a physical space hard blockage, its function value is a penalty coefficient much greater than 1. If the value is 1, then the value is 1; otherwise, the value is 1. This formula can quantitatively reflect the time matching degree and physical blockage risk of each critical path, and the weighted summation can be used to obtain the overall collaborative pressure index of the system.

[0027] The strategy generation unit in the dynamic scheduling decision module generates a non-empty and mutually exclusive set of candidate scheduling schemes based on the if-condition-then-action rule base when the collaborative pressure index exceeds the limit. For example, if the coke car is delayed in its arrival and the coke pusher has completed pushing, the rule base may trigger a scheme such as the coke pusher waiting in place or the coke pusher going to the next furnace number to push coke.

[0028] The multi-objective evaluation unit is coupled to the strategy scheme generation unit to perform multi-dimensional performance simulations on each candidate scheme in the set and calculate its comprehensive utility score for total ranking. The objectives evaluated by the multi-objective utility function include at least time objectives, energy consumption objectives, and stability objectives. The evaluation value of the time objective is based on the predicted total cycle time after the scheme is executed, which may be the total coking cycle, total waiting time, etc. The evaluation value of the energy consumption objective is based on the estimated total energy consumption of all additional vehicle travel and actions caused by the scheme, which may be the additional vehicle travel distance, the number of motor start-stops, etc. The evaluation value of the stability objective is based on the total number of state switching times required by the scheme for each vehicle to deviate from its preset standard operating sequence, which may be the number of vehicle operation sequence adjustments, furnace number switching times, etc.

[0029] Overall utility value Calculated using the following formula: ,in The first one in the current candidate scheduling scheme set One candidate scheduling scheme; To optimize the number of objectives; To optimize the target sequence number; Candidate scheduling schemes In the The original evaluation values ​​on each optimization objective; and The first one in the current candidate scheduling scheme set is the first one. Minimum and maximum evaluation values ​​for each optimization objective; For the first The optimization objective is for the candidate scheduling scheme Dynamic weights; For the first The optimization objective is for the candidate scheduling scheme The dynamic weights; the above weights depend on the performance of the scheme itself on each objective; The summation index is used to iterate through the first to last index in the exponent denominator. The optimization objectives are calculated and the corresponding dynamic weights are summed.

[0030] Dynamic weights Obtain it using the following formula: ,in yes The normalized value is . ; Candidate scheduling schemes In the The normalized value on the optimization objective is determined according to the... The same normalization rules are used to determine this; It is a sensitivity parameter greater than zero, used to control how sensitive the weights are to differences in target performance; when When the weight is large, the weight is more sensitive to the poorly performing targets and will be given a higher weight, prompting the system to prioritize improving the poorly performing targets. Through this dynamic weighted multi-objective evaluation mechanism, the system can select the scheduling scheme that performs most balanced across multiple dimensions and has the highest overall benefits.

[0031] Third embodiment: Based on the first and second embodiments, this embodiment further describes the specific process of the handshake confirmation protocol in the system monitoring and alarm module and the scheduling instruction distribution and execution module.

[0032] This system also includes a system monitoring and alarm module, which is used to monitor in real time the decision cycle of the dynamic scheduling decision module, the command response success rate of the scheduling command distribution and execution module, and the network communication status. Decision cycle monitoring ensures that the scheduling system can respond to changes in working conditions in a timely manner; command response success rate and network communication status monitoring ensure that commands can be reliably transmitted and executed. The system monitoring and alarm module establishes a multi-level alarm mechanism. When it detects equipment communication timeout, persistently high collaborative pressure index, or multiple rejections of scheduling commands, it sends audible and visual alarms, SMS notifications, or automatically generates work order information to the administrator according to the preset level. For example, when the coke tanker controller fails to confirm the scheduling command three times in a row, the system will trigger an emergency alarm and automatically generate a maintenance work order.

[0033] The specific process of the handshake confirmation protocol unit in the scheduling instruction distribution and execution module is as follows: Command Broadcast: After the dynamic scheduling decision module generates the optimal scheduling command, the scheduling command distribution and execution module broadcasts the new scheduling command to all the four relevant vehicle controllers via the industrial Ethernet bus. The four vehicle controllers are the SCP integrated machine controller, the coke quenching vehicle controller, the smoke guiding vehicle controller, and the coke tank vehicle controller.

[0034] Condition readiness confirmation: After receiving a dispatch instruction, each vehicle controller will immediately assess whether its current state meets the conditions for executing the instruction. For example, the coke tanker controller will check its own position, whether it has coke, and whether it is in a fault state. If the conditions are met, it will return a condition readiness confirmation signal to the dispatch instruction distribution and execution module.

[0035] Signal collection and judgment: The dispatch instruction distribution and execution module waits for and collects condition ready confirmation signals returned by all relevant vehicle controllers; the system will set a timeout period, and if all necessary confirmation signals are not received within the specified time, the response is considered a failure.

[0036] Final execution command: The dispatch command distribution and execution module sends the final execution command to each controller only after receiving confirmation signals that all necessary vehicles are ready.

[0037] Execution feedback: After receiving the execution command and starting execution, each vehicle controller will send execution status feedback to the scheduling command distribution and execution module; when the command is completed, it will send execution completion feedback.

[0038] Anomaly Handling: If a vehicle returns a rejection signal or fails to respond within a timeout period, the dispatch instruction distribution and execution module will cancel the current dispatch and immediately report the anomaly to the system monitoring and alarm module, which will trigger the corresponding alarm mechanism. At the same time, the dynamic dispatch decision module will reassess the current operating conditions and generate a new dispatch plan.

[0039] Through the aforementioned handshake confirmation protocol, this system ensures the reliable execution of scheduling instructions, avoids collaborative operation failures caused by information asymmetry or vehicle status mismatch, and greatly improves the stability and safety of the coke production process.

[0040] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A fully automated dynamic scheduling and control system for four main coke production vehicles, used to coordinate and control the automated operation of the four main vehicles: an SCP integrated machine, a coke quenching car, a flue gas guide car, and a coke tank car, characterized in that... include: The vehicle status perception module, based on the encoders, photoelectric switches and travel limit sensors on the four main vehicle bodies and the track, collects real-time operating data on the vehicle's absolute position, alignment with furnace number, movement trolley limit status and travel inverter status. The collaborative status prediction module is used to receive data from the vehicle status perception module, integrate historical operating data and real-time operating conditions, predict the preparation time of preceding and subsequent vehicles on key process paths through time series analysis, and calculate the collaborative pressure index that reflects the system's collaborative efficiency and congestion risk. The collaborative state prediction module includes: a time prediction unit, used to predict the completion time of the preceding vehicle action and the preparation time of the subsequent vehicle for the collaborative path from the completion of coke pushing to the coke tank car being positioned and receiving coke, and from the dust removal of the smoke guide car to the closing of the coke blocking car; and a risk field strength calculation unit, coupled to the time prediction unit, used to receive the predicted time difference of each collaborative path, and to calculate the system-level collaborative pressure index by combining the process criticality weight of the path, the ideal time window, and the physical blockage state. The collaborative path is based on the coke production process and is a pre-defined sequence of operations in the system that is completed by two or more vehicles in a specific order. The collaborative pressure index Calculated using the following formula: ,in The total number of critical collaborative paths defined in the system; For the first The inherent risk weight of each collaborative path is pre-set based on the degree of impact of path blockage on overall production. For the first Collaborative paths in time The predicted time difference is the estimated time of readiness of the subsequent vehicle minus the estimated time of readiness of the preceding vehicle. For the first The ideal time difference between the collaborative paths is the optimal time interval for achieving smooth collaboration; For the first The time difference tolerance standard deviation of each collaborative path is used to define the flexible range of time matching; For the first The feasibility impulse function of the cooperative path, when the system state vector When the path is indicated to have a physical space hard blockage, its function value is a penalty coefficient much greater than 1. Otherwise, it is 1; The dynamic scheduling decision module continuously monitors the collaborative pressure index. When it exceeds the safety threshold for dynamic adjustment of production cycle time, the strategy engine is activated. Based on the rule base, candidate solutions such as insertion waiting, step skipping, and path replanning are generated. Through multi-objective utility function evaluation and scoring that integrates time, energy consumption, and system stability, the globally optimal scheduling instruction with the highest comprehensive utility value is selected. The scheduling instruction distribution and execution module is used to parse and encapsulate the optimal scheduling instructions into a standardized instruction set that conforms to the controller protocol, and distribute them concurrently via the industrial Ethernet bus. It monitors the execution status and confirms the agreement through a handshake protocol to ensure that the relevant vehicles reach a consensus before starting collaborative operations.

2. The fully automatic dynamic scheduling and control system for four coke production vehicles according to claim 1, characterized in that: The dynamic scheduling decision module includes: The strategy scheme generation unit is used to dynamically combine basic scheduling strategies based on the rule base to generate a non-empty and mutually exclusive set of candidate scheduling schemes when the collaborative pressure index exceeds the limit. The multi-objective evaluation unit, coupled to the strategy scheme generation unit, is used to perform multi-dimensional performance simulation on each candidate scheduling scheme in the set and calculate its comprehensive utility score for total order ranking.

3. The fully automatic dynamic scheduling and control system for four coke production vehicles according to claim 2, characterized in that: The objectives evaluated by the multi-objective utility function include at least time objectives, energy consumption objectives, and stability objectives; wherein the evaluation value of the time objective is based on the predicted total cycle time after the scheme is executed; and the evaluation value of the energy consumption objective is based on the estimated total energy consumption of all additional vehicle travel and actions caused by the scheme. The evaluation value of the stability target is based on the total number of state transitions required by the scheme for each vehicle to deviate from its preset standard operating sequence.

4. The fully automatic dynamic scheduling and control system for four coke production vehicles according to claim 3, characterized in that: The overall utility value Calculated using the following formula: ,in The first one in the current candidate scheduling scheme set One candidate scheduling scheme; To optimize the number of objectives; To optimize the target sequence number; Candidate scheduling schemes In the The original evaluation values ​​on each optimization objective; and The first one in the current candidate scheduling scheme set is the first one. Minimum and maximum evaluation values ​​for each optimization objective; For the first The optimization objective is for the candidate scheduling scheme Dynamic weights; For the first The optimization objective is for the candidate scheduling scheme Dynamic weights, The summation index is used to iterate through the first to last index in the exponent denominator. The optimization objectives are calculated and the corresponding dynamic weights are summed.

5. The fully automatic dynamic scheduling and control system for four coke production vehicles according to claim 4, characterized in that: The dynamic weight Obtain it using the following formula: ,in yes The normalized value is . ; Candidate scheduling schemes In the The normalized value on the optimization objective is determined according to the... The same normalization rules are used to determine this; It is a sensitivity parameter greater than zero, used to control how sensitive the weights are to differences in target performance.

6. The fully automatic dynamic scheduling and control system for four coke production vehicles according to claim 1, characterized in that: It also includes a system monitoring and alarm module, which is used to monitor in real time the decision cycle of the dynamic scheduling decision module, the instruction response success rate of the scheduling instruction distribution and execution module, and the network communication status; The system monitoring and alarm module establishes a multi-level alarm mechanism. When it detects that the device communication timeout, the collaborative pressure index is running at a high level for a long time, or the scheduling command is rejected multiple times, it sends audible and visual alarm information, SMS or work order alarm information to the administrator according to the preset level.

7. The fully automatic dynamic scheduling and control system for four coke production vehicles according to claim 1, characterized in that: The specific process of the handshake confirmation protocol in the scheduling instruction distribution and execution module is as follows: after broadcasting a new scheduling instruction, wait for and collect condition-ready confirmation signals returned by all relevant vehicle controllers; only after receiving all necessary confirmation signals will the final execution command be sent to each controller; if any vehicle returns a rejection signal or fails to respond within a timeout period, the scheduling will be canceled and an anomaly will be reported.

Images