Concrete production task dynamic scheduling method and system considering solid waste utilization rate

Abstract

The present application belongs to the technical field of industrial production management, and specifically relates to a concrete production task dynamic scheduling method and system considering solid waste utilization rate, comprising the following steps: through an industrial internet of things system deployed in a production site, real-time collection of raw material moisture content and output rate at a pretreatment production line end, real-time inventory at an intermediate inventory end, and a to-be-scheduled order set data at an order demand end; based on collected historical moisture content sequence and output rate sequence, gradient boosting decision tree algorithm is used to predict raw material moisture content deviation and predicted output rate in a future scheduling period as scheduling boundary constraints. Through a rheological matching mechanism and urgency linkage strategy, the present application significantly improves solid waste utilization rate and production efficiency under the premise of ensuring that the rheological performance of concrete meets the standards.

Description

Technical Field

[0001] This invention belongs to the field of industrial production management technology, specifically relating to a dynamic scheduling method and system for concrete production tasks that takes into account the utilization rate of solid waste. Background Technology

[0002] In the field of industrial solid waste resource utilization, especially in the process of preparing concrete using modified phosphogypsum, the instability of raw material quality is a core challenge restricting large-scale production. After pretreatment processes such as neutralization and modification, the moisture content, pH, and activity index of phosphogypsum are not constant but exhibit nonlinear dynamic fluctuations depending on the batch of raw materials. This fluctuation has vastly different impacts on concrete of different grades; high-grade concrete is extremely sensitive to even minor fluctuations in raw materials, while low-grade concrete has a higher tolerance. However, existing scheduling systems often employ fixed proportions or simple first-come, first-served strategies, neglecting the matching relationship between the real-time status of raw materials and order characteristics. This can easily lead to quality incidents in high-grade products due to raw material fluctuations.

[0003] On the other hand, pretreatment production lines typically operate in a continuous flow mode, while downstream mixing plants operate based on discrete orders, resulting in a natural mismatch between their production and consumption rhythms. Existing scheduling methods lack effective prediction and linkage mechanisms for intermediate warehouse inventory levels. When upstream output is too rapid and inventory is high, if the scheduling system still schedules orders with low solid waste content, it can easily lead to intermediate warehouse overflow and force the production line to shut down. Conversely, if high-content orders are scheduled when inventory is insufficient, it will lead to material shortages and waiting times, severely impacting production continuity and efficiency.

[0004] Furthermore, traditional production scheduling often employs standard non-dominated sorting genetic algorithms, whose parameters, such as crossover and mutation probabilities, are typically fixed. Faced with the rapidly changing urgency of inventory and the difficulty of matching raw materials in solid waste treatment scenarios, algorithms with fixed parameters cannot adaptively adjust their search strategies. They struggle to quickly find the optimal solution that balances maximizing solid waste disposal with on-time order delivery within minute-level time windows, often leading to scheduling schemes that get stuck in local optima and fail to meet actual production needs. Summary of the Invention

[0005] This invention provides a method and system for dynamic scheduling of concrete production tasks that takes into account the utilization rate of solid waste, in order to solve the technical problems in the prior art that lead to unstable concrete quality due to ignoring the fluctuation of raw material quality, and that lead to silo overload or material shortage shutdowns due to the mismatch between production and consumption rhythms.

[0006] In a first aspect, the present invention provides a dynamic scheduling method for concrete production tasks that takes into account the utilization rate of solid waste, comprising the following steps: By deploying an industrial IoT system on the production site, real-time data can be collected on the raw material moisture content and output rate at the pre-processing production line, the real-time inventory at the intermediate inventory level, and the set of pending production orders at the order demand level. Based on the collected historical moisture content sequence and output rate sequence, the gradient boosting decision tree algorithm is used to predict the raw material moisture content deviation and the predicted output rate in the next scheduling cycle, which serve as scheduling boundary constraints. Based on the rheological sensitivity coefficient of the corresponding order number and the deviation of raw material moisture content, a suitable evaluation index for the rheological matching of solid waste tasks is constructed. Based on real-time inventory, optimal safety stock level and current output rate, a solid waste disposal urgency index evaluation index is constructed. By integrating the solid waste task rheological matching suitability evaluation index and the solid waste disposal urgency index evaluation index into the non-dominated sorting genetic algorithm, and using the production status index to dynamically adjust the crossover probability of the algorithm, the optimal task sequence that balances maximizing the amount of solid waste disposal and minimizing the total delay time is obtained, thereby realizing the dynamic scheduling of concrete production tasks.

[0007] Its effects are as follows: By collecting multi-source data in real time through the Internet of Things and combining it with machine learning to predict the state of raw materials, this invention innovatively constructs rheological matching suitability and disposal urgency index as evaluation indicators, and then adaptively adjusts the non-dominated sorting genetic algorithm to optimize production scheduling. This achieves a precise mapping of the physical state of the concrete production site to scheduling decision variables. Under the premise of ensuring the stable quality of concrete of different grades, it significantly improves the utilization rate of solid waste resources and the collaborative efficiency of continuous production.

[0008] Furthermore, an evaluation index for the suitability of solid waste task rheology matching is constructed, including: Get the rheological sensitivity coefficient corresponding to each order number ,as well as The deviation of solid waste moisture content from the standard value at any given time ; The suitability of solid waste task rheology matching is calculated using the following formula. : ; In the formula, For adjustment coefficients, The natural constant; rheological sensitivity coefficient It is a dimensionless constant obtained by measuring and normalizing the slump change rate of concrete of different grades under moisture content fluctuations; moisture content deviation This is the absolute value of the difference between the real-time moisture content and the standard moisture content; this indicator is used to assess how well an order adapts to fluctuations in raw material prices.

[0009] Its effect is that it quantifies the suitability of solid waste task rheological matching through a specific logarithmic composite relationship. Compared with subjective experience judgment, this formula scientifically integrates the physical sensitivity differences of different grades of concrete to water-cement ratio and the absolute deviation of moisture content. It can accurately identify the adaptability of orders to raw material fluctuations, so that the scheduling system can automatically shield high-risk orders and prioritize low-sensitivity orders when faced with large fluctuations in raw material moisture content, thus avoiding the risk of quality scrapping of high-grade products from the source.

[0010] Furthermore, an evaluation index for the urgency of solid waste disposal is constructed, including: Get current real-time inventory Optimal safety stock level Current output rate and maximum design output rate ; The solid waste disposal urgency index is calculated using the following formula. : ; In the formula, It is a natural constant. This indicates the larger of 0 and the value in parentheses; the solid waste disposal urgency index is used to assess the urgency of the system's need to dispose of solid waste. When the inventory exceeds the optimal water level and the output rate is high, the solid waste disposal urgency index value increases.

[0011] Its effect is as follows: by constructing a solid waste disposal urgency index using nonlinear logarithmic and ratio relationships, it overcomes the shortcomings of traditional scheduling that lacks forward-looking prediction of intermediate warehouse water levels. This index can keenly capture and quantify dangerous physical states such as high inventory and excessively rapid output, providing accurate numerical basis for the system to assess the true urgency of solid waste disposal, effectively preventing the pretreatment production line from being forced to shut down due to warehouse overflow or downstream material shortages, and ensuring the smooth and continuous production and disposal rhythm of upstream and downstream.

[0012] Furthermore, the gradient boosting decision tree algorithm is used to predict the raw material moisture content deviation and the predicted output rate within a future scheduling cycle, including: Extract the moisture content sequence, production rate sequence, pretreatment reactor temperature, and stirring current within the historical time window as input features; Supervised learning samples are constructed by slicing historical data into sliding windows, and a robust loss function is used to train a gradient boosting decision tree regression model. Output the predicted moisture content deviation and predicted output rate for the future scheduling cycle to eliminate control lag.

[0013] Its effect is as follows: by adopting a gradient boosting decision tree algorithm that combines a robust loss function and introducing a sliding window technique, it solves the serious lag problem of traditional feedback control. It can perceive and accurately output the fluctuation trend of raw material moisture content and output rate in advance within a short period of time based on auxiliary variables, so that production scheduling is transformed from passive remediation after the fact to proactive prevention before the fact, ensuring that the generated scheduling plan will not fail due to sudden changes in the state of raw materials at the production site.

[0014] Furthermore, the crossover probability of the algorithm is dynamically adjusted using production status indicators, including: Based on the solid waste disposal urgency index and the average matching suitability of the current pending order set. The adaptive crossover probability is calculated using the following formula. : ; In the formula, Based on the basic crossover probability, This is the upper limit of probability. This is the dynamic gain coefficient. The average of the sum of the suitability scores for solid waste tasks in all pending orders. The function is used to ensure that the final probability does not exceed the upper probability limit. The urgency index for solid waste disposal.

[0015] Its effect is as follows: It establishes a dynamic response mechanism based on the adaptive adjustment of the crossover probability of the genetic algorithm according to the system environment state, which breaks the limitation of the traditional optimization algorithm with fixed parameters, which is prone to getting trapped in local optima when facing harsh working conditions. In extreme scenarios with high inventory pressure and poor raw material matching, it can automatically increase the crossover probability to force the algorithm to jump out of local solutions, thereby improving the system's optimization robustness and decision-making efficiency in complex and ever-changing production environments.

[0016] Furthermore, the optimal task sequence that balances maximizing solid waste disposal volume and minimizing total delay time is obtained, including: Construct a first objective function to maximize the amount of solid waste disposed of, which is calculated as the sum of the products of the demand volume of all orders and the actual amount of solid waste mixed in. A second objective function is constructed to minimize the total delay time, which is calculated as the sum of the positive values ​​of the difference between the completion time and the delivery deadline for all orders.

[0017] Furthermore, the collected data also includes: obtaining a set of production orders to be scheduled from the enterprise resource planning system, and extracting key attributes including demand volume, delivery deadline, concrete strength grade and the maximum allowable amount of solid waste for that grade; the moisture content of raw materials at the pretreatment production line end is obtained through a microwave online moisture meter, and the output rate is obtained through an electronic belt scale.

[0018] Furthermore, after achieving dynamic scheduling of concrete production tasks, it also includes: The selected task sequence is broken down into specific batching instructions and sent to the mixing plant controller; The scheduling steps are re-executed at preset time intervals. If the newly generated plan is better than the currently executed plan by more than a preset proportion, the order insertion and re-sorting are triggered.

[0019] Furthermore, after realizing the dynamic scheduling of concrete production tasks, it also includes: after the algorithm outputs the Pareto optimal solution set, calculating the congestion distance of each solution, and prioritizing the solution with a large congestion distance and a total delay time less than a preset threshold as the final solution.

[0020] Secondly, the present invention provides a dynamic scheduling system for concrete production tasks that takes into account the utilization rate of solid waste, including a memory and a processor. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned dynamic scheduling method for concrete production tasks that takes into account the utilization rate of solid waste is implemented.

[0021] The beneficial effects are as follows: This invention constructs a closed-loop scheduling system capable of real-time evaluation of raw material and order matching and sensing inventory pressure. By introducing the suitability of solid waste task rheological matching and the urgency index of solid waste disposal, it successfully maps the physical state of the production site into the decision variables of the scheduling algorithm. Combined with an improved adaptive genetic algorithm, it solves the problems of fixed parameters and poor adaptability of traditional algorithms. This invention can not only effectively avoid the quality risks caused by raw material fluctuations, but also maximize the utilization of solid waste resources while ensuring production continuity, significantly improving the economic and environmental benefits of enterprises. Attached Figure Description

[0022] Figure 1 This is a flowchart of the dynamic scheduling method for concrete production tasks that takes into account the utilization rate of solid waste, as described in this invention.

[0023] Figure 2 This is a comparison diagram of the distribution of the multi-objective scheduling optimization solution set of the present invention and the solution set of the prior art.

[0024] Figure 3 This is a distribution diagram showing the matching between the solid waste raw material characteristics and order rheological sensitivity of the present invention.

[0025] Figure 4 This is a graph showing the dispersion of product strength under different solid waste substitution rates according to the present invention. Detailed Implementation

[0026] 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, not all, of the embodiments of the present invention. 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.

[0027] An embodiment of the dynamic scheduling method for concrete production tasks that considers solid waste utilization provided by this invention: like Figure 1 As shown, the dynamic scheduling method for concrete production tasks considering solid waste utilization includes the following steps: S1, through an industrial IoT system deployed on the production site, collects in real time data on the moisture content and output rate of raw materials at the pre-processing production line, the real-time inventory at the intermediate inventory level, and the set of pending production orders at the order demand level.

[0028] Specifically, the data collection dimensions cover the pre-processing production line, intermediate inventory, and order demand, with a sampling frequency set at 1Hz to ensure data real-time performance.

[0029] At the pretreatment production line, the moisture content data of the modified phosphogypsum is read in real time using a microwave online moisture meter installed on a belt conveyor. The microwave moisture meter calculates the internal moisture content of the material by emitting microwave signals and receiving the attenuation signals after they penetrate the material. Its data source is accurate and reliable. Simultaneously, a high-precision electronic belt scale installed below the conveyor belt is used to obtain the real-time production rate. The pH level is monitored in real time using an online pH sensor.

[0030] At the intermediate storage end, the current real-time inventory level is obtained by using radar level gauges installed on the top of the solid waste intermediate silo, which utilize the principle of echo ranging. .

[0031] On the order demand side, the production schedule order table is read from the enterprise's ERP system database via a data interface, and key fields are extracted, including: demand volume. Delivery deadline Concrete strength grade and the maximum allowable solid waste content according to the process specifications for that grade. .

[0032] By collecting key physical quantities in real time throughout the entire production process, information silos are broken down, providing a real and accurate data foundation for subsequent precise scheduling and avoiding errors caused by relying on manual experience for estimation.

[0033] S2, based on the collected historical moisture content sequence and output rate sequence, uses the gradient boosting decision tree algorithm to predict the raw material moisture content deviation and the predicted output rate in the next scheduling cycle, as scheduling boundary constraints.

[0034] Specifically, in order to eliminate control lag, this step uses a gradient boosting decision tree algorithm to predict the state of raw materials in the short term.

[0035] First, feature engineering is constructed: extract the time series of moisture content and production rate over the past 2 hours, as well as auxiliary variables such as pretreatment reactor temperature and stirring current as input features.

[0036] Secondly, model training: A sliding window is used to slice the historically stored data to construct input-output pairs. A robust loss function is then used to train the gradient boosting decision tree regression model. The robust loss function combines the advantages of mean squared error and absolute error, is insensitive to occasional noise from sensor data, and ensures the model's robustness.

[0037] Finally, the predicted output: the model outputs the predicted water content deviation for the next scheduling cycle. And predict the output rate.

[0038] For example, if the current moisture content is 13%, and the model predicts that the moisture content will rise to 13.5% within the next 30 minutes based on the trend, it will output a deviation signal in advance. These predicted values ​​will serve as constraints for subsequent optimization algorithms to prevent the scheduling scheme from failing due to sudden changes in raw materials after it has been generated.

[0039] By introducing short-term prediction based on machine learning, the system can detect raw material fluctuation trends in advance, realizing the transformation from post-event remediation to pre-event prevention and effectively eliminating the lag in feedback control.

[0040] S3. Based on the rheological sensitivity coefficient of the corresponding order number and the deviation of raw material moisture content, construct the solid waste task rheological matching suitability evaluation index, and based on the real-time inventory, the optimal safety stock level and the current output rate, construct the solid waste disposal urgency index evaluation index.

[0041] This step is the core component. It transforms physical laws into scheduling criteria through explicit mathematical formulas, ensuring the objectivity and rationality of the evaluation results.

[0042] (1) Constructing the suitability of solid waste task rheological matching

[0043] Different grades of concrete have different sensitivities to fluctuations in the moisture content of raw materials. High-grade concrete has a low water-cement ratio and is extremely sensitive to water, while low-grade concrete has a high tolerance.

[0044] First, determine the rheological sensitivity coefficient. This data comes from laboratory measurements: determining the moisture content fluctuation of concrete of different grades. The rate of change of slump at time, and normalize the rate of change to A dimensionless constant. For example, let C60 be... C30 .

[0045] Secondly, calculate the moisture content deviation. Assume the standard process moisture content If the predicted real-time moisture content The absolute deviation value is .

[0046] Set adjustment coefficient Substitute into the relation:

[0047] Calculation example: Assuming the current time-to-date prediction of raw material moisture content deviation .

[0048] For high-grade order A, : Index Term ; denominator ; For order B with a lower grade, : Denominator ; .

[0049] 0.152 is much greater than 0.056. The calculation results show that when the moisture content of the raw materials fluctuates greatly, the system scores the low-grade orders significantly higher than the high-grade orders.

[0050] By calculating and ranking these indicators, the system can automatically filter out high-risk orders and prioritize low-sensitivity orders, thus avoiding quality incidents from the source by utilizing physical laws.

[0051] (2) Constructing a solid waste disposal urgency index

[0052] This indicator is used to measure inventory pressure; it provides real-time data: current inventory. tons, optimal safety stock tons, current production rate tons / hour, maximum designed output rate tons / hour.

[0053] Substitute the relation:

[0054] Calculation example: Inventory ratio item: Output ratio index: ; Exponentiation part: Internal logarithmic item: Inventory excess rate Logarithmic terms: Final result: .

[0055] Comparison: If the inventory is only 400 tons, then The logarithmic terms are Exponentiation part The final result is less than 1, indicating that it is not urgent.

[0056] This indicator can keenly capture the dangerous situation of high inventory and rapid production. The quantitative value truly reflects the current urgency of consumption and provides a basis for algorithm adjustment.

[0057] S4 integrates the solid waste task rheological matching suitability evaluation index and the solid waste disposal urgency index evaluation index into the non-dominated sorting genetic algorithm. By dynamically adjusting the crossover probability of the algorithm using production status indicators, the optimal task sequence that balances maximizing solid waste disposal volume and minimizing total delay time is obtained, thereby realizing the dynamic scheduling of concrete production tasks.

[0058] Specifically, this step feeds back the aforementioned metrics to the non-dominated sorting genetic algorithm, forming a closed loop.

[0059] The optimization objectives are set as follows: the first objective is to maximize the amount of solid waste disposed of. Second objective: Minimize total delay time .

[0060] To avoid the algorithm getting trapped in local optima, the crossover probability is dynamically adjusted. Set basic parameters: basic crossover probability Upper probability limit Dynamic gain coefficient ; Calculate the average of the sum of the solid waste task rheology matching suitability of all pending orders. Assuming there are 10 orders pending, the calculated... The average value is 0.1; referring to the aforementioned urgency index. Substitute into the relation:

[0061] Calculation example: Incremental part Basic superposition Take the minimum value: .

[0062] The calculation results show that under the harsh conditions of extremely high inventory pressure and poor raw material matching, the crossover probability is automatically adjusted to the upper limit of 0.95. This means that the algorithm will perform high-intensity population crossover operations, forcibly break out of the original local solution, and try to find an unconventional scheduling sequence.

[0063] After the algorithm completes execution, it outputs the Pareto optimal solution set. The system calculates the congestion distance and automatically selects the solution with the largest congestion distance and the longest total delay time. The solution is within an acceptable range. Finally, the selected optimal task sequence is parsed into specific batching instructions and sent to the mixing plant's PLC control system for execution.

[0064] By adaptively adjusting algorithm parameters, the system demonstrates strong robustness in the face of complex and ever-changing production environments, ensuring both production safety and balancing quality and efficiency, thus achieving globally optimal scheduling.

[0065] like Figure 2 As shown in the figure, this is a comparison of the Pareto front for multi-objective optimization. With the total delay time as the horizontal axis and the solid waste disposal volume as the vertical axis, it compares and shows the optimal solution distribution of the present invention and the prior art. The solution set of the prior art is concentrated in the high delay-low disposal or low delay-very low disposal range and is sparsely distributed, easily trapped in local optima. In contrast, the solution set of the present invention is generally shifted to the upper left, with a shorter delay time and a higher disposal volume for the same solid waste disposal volume, and a wider and more uniform distribution. This verifies that the improved algorithm can effectively balance the dual objectives of maximizing solid waste disposal and minimizing on-time order delivery.

[0066] like Figure 3 As shown in the figure, the solid waste raw material moisture content deviation is used as the horizontal axis and the order rheological sensitivity coefficient is used as the vertical axis. The rheological matching suitability of solid waste tasks is marked by color or numerical value, showing obvious gradient distribution characteristics. Areas with small moisture content deviation and low rheological sensitivity coefficient have higher matching suitability, while high-risk areas with high deviation and high sensitivity have matching suitability close to 0. The abstract mathematical calculation of matching degree is transformed into an intuitive spatial distribution, which not only confirms the scientific nature of the rheological matching suitability evaluation index, but also provides a visual basis for the scheduling strategy of prioritizing low-sensitivity orders when raw material fluctuations occur.

[0067] like Figure 4 As shown in the figure, the solid waste replacement rate is plotted on the horizontal axis and the product strength dispersion is plotted on the vertical axis. The results compare the application effects of the present invention with those of the traditional scheduling method. Under the traditional method, the product strength dispersion increases rapidly with the increase of the solid waste replacement rate, and the quality fluctuation of high-grade concrete is particularly obvious. However, under the same solid waste replacement rate, the product strength dispersion of the present invention is significantly lower, and the upward trend with the increase of the replacement rate is gradual. Even with a high solid waste content, the dispersion can still be controlled within the allowable range of the process. This proves that the rheological matching mechanism of the present invention can effectively ensure the stability of concrete product quality while greatly improving the utilization rate of solid waste.

[0068] An embodiment of the dynamic scheduling system for concrete production tasks that considers solid waste utilization provided by this invention: The dynamic scheduling system for concrete production tasks that considers solid waste utilization includes a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned dynamic scheduling method for concrete production tasks that considers solid waste utilization is implemented.

[0069] The dynamic scheduling system for concrete production tasks that takes into account the utilization rate of solid waste also includes other components well known to those skilled in the art, such as communication interfaces. Their settings and functions are known in the art and will not be described in detail here.

[0070] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0071] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A dynamic scheduling method for concrete production tasks considering solid waste utilization, characterized in that, Includes the following steps: By deploying an industrial IoT system on the production site, real-time data can be collected on the raw material moisture content and output rate at the pre-processing production line, the real-time inventory at the intermediate inventory level, and the set of pending production orders at the order demand level. Based on the collected historical moisture content sequence and output rate sequence, the gradient boosting decision tree algorithm is used to predict the raw material moisture content deviation and the predicted output rate in the next scheduling cycle, which serve as scheduling boundary constraints. Based on the rheological sensitivity coefficient of the corresponding order number and the deviation of raw material moisture content, a suitable evaluation index for the rheological matching of solid waste tasks is constructed. Based on real-time inventory, optimal safety stock level and current output rate, a solid waste disposal urgency index evaluation index is constructed. The evaluation index of suitability of solid waste task rheological matching and evaluation index of urgency of solid waste disposal are integrated into the non-dominated sorting genetic algorithm. The crossover probability of the algorithm is dynamically adjusted by the production status index. The optimal task sequence that balances maximizing the amount of solid waste disposal and minimizing the total delay time is obtained, so as to realize the dynamic scheduling of concrete production tasks. Construct a suitability evaluation index for solid waste task rheology matching, including: obtaining the first... rheological sensitivity coefficient corresponding to each order number ,as well as The deviation of solid waste moisture content from the standard value at any given time ; The suitability of solid waste task rheology matching is calculated using the following formula. : ； In the formula, For adjustment coefficients, The natural constant; rheological sensitivity coefficient It is a dimensionless constant obtained by measuring and normalizing the slump change rate of concrete of different grades under moisture content fluctuations; moisture content deviation This is the absolute value of the difference between the real-time moisture content and the standard moisture content; this indicator is used to assess the adaptability of an order to fluctuations in raw material prices. Construct an evaluation index for the urgency of solid waste disposal, including: Get current real-time inventory Optimal safety stock level Current output rate and maximum design output rate ; The solid waste disposal urgency index is calculated using the following formula. : ； In the formula, It is a natural constant. This indicates the larger of 0 and the value in parentheses; the solid waste disposal urgency index is used to assess the urgency of the system's need to dispose of solid waste. When the inventory exceeds the optimal level and the output rate is high, the solid waste disposal urgency index value increases.

2. The dynamic scheduling method for concrete production tasks considering solid waste utilization rate according to claim 1, characterized in that, The gradient boosting decision tree algorithm is used to predict the raw material moisture content deviation and the production rate within a future scheduling cycle, including: Extract the moisture content sequence, production rate sequence, pretreatment reactor temperature, and stirring current within the historical time window as input features; Supervised learning samples are constructed by slicing historical data into sliding windows, and a robust loss function is used to train a gradient boosting decision tree regression model. Output the predicted moisture content deviation and predicted output rate for the future scheduling cycle to eliminate control lag.

3. The dynamic scheduling method for concrete production tasks considering solid waste utilization rate according to claim 1, characterized in that, The crossover probability of the algorithm is dynamically adjusted using production status indicators, including: Based on the solid waste disposal urgency index and the average matching suitability of the current pending order set. The adaptive crossover probability is calculated using the following formula. : ； In the formula, Based on the basic crossover probability, This is the upper limit of probability. This is the dynamic gain coefficient. The average of the sum of the suitability scores for solid waste tasks in all pending orders. The function is used to ensure that the final probability does not exceed the upper probability limit. The urgency index for solid waste disposal.

4. The dynamic scheduling method for concrete production tasks considering solid waste utilization rate according to claim 1, characterized in that, The optimal task sequence that balances maximizing solid waste disposal volume and minimizing total delay time is obtained, including: Construct a first objective function to maximize the amount of solid waste disposed of, which is calculated as the sum of the products of the demand volume of all orders and the actual amount of solid waste mixed in. A second objective function is constructed to minimize the total delay time, which is calculated as the sum of the positive values ​​of the difference between the completion time and the delivery deadline for all orders.

5. The dynamic scheduling method for concrete production tasks considering solid waste utilization rate according to claim 1, characterized in that, The collected data also includes: obtaining a set of production orders to be scheduled from the enterprise resource planning system, and extracting key attributes including demand volume, delivery deadline, concrete strength grade and the maximum allowable amount of solid waste for that grade; the moisture content of raw materials at the pretreatment production line end is obtained through a microwave online moisture meter, and the output rate is obtained through an electronic belt scale.

6. The dynamic scheduling method for concrete production tasks considering solid waste utilization rate according to claim 1, characterized in that, After achieving dynamic scheduling of concrete production tasks, it also includes: The selected task sequence is broken down into specific batching instructions and sent to the mixing plant controller; The scheduling steps are re-executed at preset time intervals. If the newly generated plan is better than the currently executed plan by more than a preset proportion, the order insertion and re-sorting are triggered.

7. The dynamic scheduling method for concrete production tasks considering solid waste utilization rate according to claim 1, characterized in that, After achieving dynamic scheduling of concrete production tasks, the algorithm also includes: calculating the congestion distance of each solution after outputting the Pareto optimal solution set, and prioritizing the solution with a large congestion distance and a total delay time less than a preset threshold as the final solution.

8. A dynamic scheduling system for concrete production tasks considering solid waste utilization, characterized in that, It includes a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the dynamic scheduling method for concrete production tasks considering solid waste utilization rate as described in any one of claims 1-7 is implemented.

Images