Agricultural machinery operation plan adjustment method considering weather dynamic change

Abstract

The invention discloses an agricultural machinery operation plan adjustment method considering dynamic weather change, belongs to the field of agricultural machinery scheduling, and aims to solve the problems of frequent failure, cost increase and agricultural time delay in actual execution of an operation plan due to the fact that the dynamic weather change is not fully considered in an existing agricultural machinery scheduling method. The method comprises the steps that firstly, an initial scheduling model with the minimum total cost as the target is constructed based on predicted weather, the model is fused with a farmland operation time window and path continuity constraint, an improved bat algorithm is adopted for solving, and an initial optimal scheme containing the agricultural machinery operation path, sequence and time is output; in the operation process, the system monitors weather in real time, and when an original time window loses efficacy or a path is not feasible due to changes, dynamic adjustment is automatically triggered. And then, based on the real-time position of the agricultural machine and the updated weather information, rapidly reconstructing and solving a dynamic optimization model, and outputting an optimal adjustment scheme of the remaining tasks.

Description

Technical Field

[0001] This invention relates to the field of agricultural machinery scheduling, specifically to a method for adjusting agricultural machinery operation plans that takes into account dynamic weather changes, and more particularly to an initial planning and dynamic adjustment optimization method for agricultural machinery operations with the goal of minimizing total scheduling costs. Background Technology

[0002] With the rapid development of modern science and technology, new-generation information technologies such as the Internet of Things, big data, and artificial intelligence have been widely applied in the agricultural field, promoting the in-depth development of smart agriculture. In the field of agricultural machinery operation scheduling, rationally planning agricultural machinery operation paths and times can effectively improve operation efficiency and reduce operation costs, and has become a key link in realizing refined and intelligent agricultural management.

[0003] Agricultural machinery operations, as a typical outdoor production activity, are highly dependent on weather conditions. Different weather conditions (such as rain, strong winds, and frost) directly affect the working efficiency, operation quality, and even the feasibility of agricultural machinery. For example, rain may cause slippery soil in farmland, affecting the speed of agricultural machinery and operational safety; continuous rainy weather will significantly shorten the window of suitable operation time. Therefore, when formulating agricultural machinery operation plans, the dynamic impact of weather factors must be fully considered to ensure the feasibility of the plan and the requirements of agricultural timing.

[0004] Currently, most existing agricultural machinery scheduling methods are optimized based on static or deterministic environmental assumptions, meaning that parameters such as weather conditions and operating times are assumed to remain constant during the planning stage. While these methods can achieve good optimization results under ideal conditions, in practical applications, due to the unpredictability and dynamic changes in weather, pre-planned static scheduling schemes are often difficult to implement effectively. When sudden weather changes occur, the original plan may be unable to continue due to the failure of the time window or the infeasibility of the route, leading to operation interruptions, idle agricultural machinery, increased costs, and even missing critical farming seasons, seriously affecting agricultural production efficiency.

[0005] In recent years, although some research has begun to focus on scheduling problems under uncertain environments, in the field of agricultural machinery operations, how to systematically integrate dynamic weather forecast information into scheduling models and design efficient and adaptive dynamic adjustment mechanisms remains a pressing technical challenge. Existing dynamic adjustment methods mostly focus on emergency response, lack coordination optimization with the initial plan, and have high computational complexity, making it difficult to meet the dual requirements of real-time performance and economy in large-scale, multi-machine collaborative operation scenarios.

[0006] Therefore, there is an urgent need in this field for a method for generating and dynamically adjusting agricultural machinery operation plans that can deeply integrate dynamic weather change information, take into account the economy and robustness of scheduling schemes, and have efficient solution capabilities, in order to cope with the complex and ever-changing environmental challenges in actual agricultural production and achieve efficient, economical, and reliable scheduling of agricultural machinery resources. Summary of the Invention

[0007] To address the problems of frequent failures, increased costs, and delays in agricultural operations caused by existing agricultural machinery scheduling methods that do not adequately consider dynamic weather changes, this invention provides a method for adjusting agricultural machinery operation plans that takes into account dynamic weather changes.

[0008] The present invention discloses a method for adjusting agricultural machinery operation plans that takes into account dynamic weather changes, comprising the following steps:

[0009] S1. Construct an initial scheduling optimization model whose objective function is to minimize the total scheduling cost, with the following constraints:

[0010] Constraints on farmland operation time windows determined based on forecast weather information;

[0011] Continuity constraints on agricultural machinery operation paths;

[0012] S2. The improved bat heuristic algorithm is used to solve the initial scheduling optimization model and output the initial optimal scheduling scheme, which includes the operation path, operation sequence and operation time arrangement of each agricultural machine.

[0013] S3. Real-time monitoring of weather conditions. When actual weather changes are detected that cause the time window for farmland operations to fail or the driving path of agricultural machinery to become infeasible, a dynamic adjustment mechanism is triggered, and the following dynamic adjustment steps are initiated.

[0014] S4. Based on the current completed task status, real-time location information of agricultural machinery, and updated weather information, a dynamic scheduling optimization model is constructed. Its objective function is to minimize the total scheduling cost of the remaining tasks. Constraints include:

[0015] Constraints on farmland operation time windows redefined based on updated weather information;

[0016] Path continuity constraints for agricultural machinery starting from its current position;

[0017] S5. The improved bat heuristic algorithm is used to solve the dynamic scheduling optimization model, and the dynamic optimal scheduling scheme is output to guide the execution of the remaining tasks.

[0018] Preferably, in step S1, the objective function of the initial optimization model is:

[0019]

[0020] In the formula, For the total scheduling cost, , For operating costs, For path cost;

[0021] For agricultural machinery farmland The unit cost of the operation. , For agricultural machinery collection, , Collect farmland;

[0022] For agricultural machinery from position arrive The cost of driving , , For the set of location points;

[0023] As path decision variables, if agricultural machinery From position Go to If yes, then it is 1; otherwise, it is 0.

[0024] Assign decision variables to the operation: if agricultural machinery farmland If the assignment is correct, the value is 1; otherwise, it is 0.

[0025] Preferably, in step S1, the operation time window constraint is: for any farmland Its operation start time Must meet ,in This refers to the feasible operation time window determined based on forecast weather information.

[0026] Preferably, in steps S2 and S5, the improved bat heuristic algorithm adds the following mechanism to the standard bat algorithm:

[0027] Coding Design: The scheduling scheme is encoded into a farm machinery-farmland allocation vector and a farm machinery internal access order vector;

[0028] Fitness function design: fitness value The expression is: , The penalty value for violating the time window;

[0029] Local search strategy: When the local search triggering condition is met, a new solution is constructed by introducing random perturbation based on the current optimal solution.

[0030] Preferably, in step S4, the dynamic scheduling optimization model is adjusted as follows based on the initial scheduling optimization model: updating the set of farmland, the set of current locations of agricultural machinery, and the set of operation time windows.

[0031] Preferably, the solution process of the dynamic scheduling optimization model includes:

[0032] Based on the portion of the initial optimal scheduling scheme output in step S2 concerning the remaining tasks, construct a partially retained seed individual;

[0033] Perturbation is performed on the seed individuals to generate a new optimized population;

[0034] The new population is optimized using the same bat algorithm framework as in step S2.

[0035] Preferably, the method further includes outputting the operation path, operation sequence, and time schedule for each agricultural machine.

[0036] The beneficial effects of this invention are:

[0037] 1. Significantly improved the robustness and adaptability of agricultural machinery scheduling plans.

[0038] By constructing a two-layer model that combines initial optimization and dynamic optimization, this invention can systematically integrate short-term and real-time weather forecast information. This enables the scheduling scheme to not only be globally optimized based on initial meteorological conditions, but also to quickly and automatically replan the remaining tasks when the weather changes dynamically. This effectively avoids the problem of traditional static scheduling plans failing as a whole due to sudden weather changes, and ensures the continuous executability of the work plan.

[0039] 2. Effective minimization of total scheduling cost was achieved.

[0040] The model aims to minimize the total scheduling cost (including operation cost and path cost) and employs an efficient bat heuristic algorithm for solution. This method can find the economically optimal scheduling scheme while satisfying complex constraints such as all farmland operation needs, time windows, and path continuity. Compared with empirical scheduling or traditional optimization methods, it can significantly reduce fuel consumption, farm machinery idle time, and waiting time, thereby achieving practical economic benefits of cost reduction and efficiency improvement.

[0041] 3. Improved solution efficiency and solution quality under complex constraints.

[0042] For the NP-hard combinatorial optimization problem of agricultural machinery scheduling, this invention designs a bat algorithm solution framework tightly coupled with the characteristics of the problem. By designing a reasonable decoding method, incorporating a fitness function with a time window penalty mechanism, and an update strategy that combines global search and local fine-tuning, the algorithm can output high-quality, feasible, and optimized operation schemes for scheduling large-scale farmland and agricultural machinery clusters within an acceptable time, balancing the dual requirements of solution speed and scheme accuracy.

[0043] 4. Enhanced the intelligent decision-making capabilities of smart agriculture systems.

[0044] This invention provides a complete decision support solution, from model to algorithm, which can be directly integrated into a smart agriculture management platform. Based on real-time sensor data and weather forecasts, the system can automatically trigger dynamic adjustments to the scheduling plan, reducing reliance on and lag in manual intervention. This promotes the upgrade of agricultural machinery scheduling from "static planning" to "dynamic intelligent response," thereby improving the level of intelligent management in agricultural production.

[0045] 5. Possesses good scalability and potential for widespread application.

[0046] The "initial planning + dynamic response" framework and intelligent algorithm-based solution method proposed in this invention are not only applicable to weather factors, but their core ideas can also be extended to address other uncertainties and disturbances such as agricultural machinery malfunctions, changes in task priorities, and changes in farmland conditions. Therefore, this technical solution also has broad application prospects in dynamic optimization fields such as vehicle routing, logistics distribution, and personnel scheduling, beyond agriculture.

[0047] In summary, this invention, through technological innovation, establishes an optimization model at the theoretical level that is more in line with the dynamics of actual production, and provides an efficient, automatic, and economical intelligent scheduling solution at the application level. It can effectively overcome the constraints of weather uncertainty on agricultural machinery operations and provide key technical support for the stable and efficient operation of smart agriculture. Attached Figure Description

[0048] Figure 1 This is a flowchart of a method for adjusting agricultural machinery operation plans that takes into account dynamic weather changes, as described in this invention. Detailed Implementation

[0049] 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.

[0050] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0051] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the scope of the invention.

[0052] Specific Implementation Method 1: The following is combined with... Figure 1 This embodiment describes a method for adjusting agricultural machinery operation plans that takes into account dynamic weather changes, comprising the following steps:

[0053] S1. Construct an initial scheduling optimization model whose objective function is to minimize the total scheduling cost, with the following constraints:

[0054] Constraints on farmland operation time windows determined based on forecast weather information;

[0055] Continuity constraints on agricultural machinery operation paths;

[0056] S2. The improved bat heuristic algorithm is used to solve the initial scheduling optimization model and output the initial optimal scheduling scheme, which includes the operation path, operation sequence and operation time arrangement of each agricultural machine.

[0057] S3. Real-time monitoring of weather conditions. When actual weather changes are detected that cause the time window for farmland operations to fail or the driving path of agricultural machinery to become infeasible, a dynamic adjustment mechanism is triggered, and the following dynamic adjustment steps are initiated.

[0058] S4. Based on the current completed task status, real-time location information of agricultural machinery, and updated weather information, a dynamic scheduling optimization model is constructed. Its objective function is to minimize the total scheduling cost of the remaining tasks. Constraints include:

[0059] Constraints on farmland operation time windows redefined based on updated weather information;

[0060] Path continuity constraints for agricultural machinery starting from its current position;

[0061] S5. The improved bat heuristic algorithm is used to solve the dynamic scheduling optimization model, and the dynamic optimal scheduling scheme is output to guide the execution of the remaining tasks.

[0062] The following is a detailed explanation.

[0063] I. Preliminary Data Preparation and Parameter Setting

[0064] Suppose an agricultural machinery cooperative needs to cultivate 10 scattered plots of farmland (numbered F1-F10) during the busy farming season and owns 3 agricultural machines of the same model (numbered M1-M3). The dispatch center obtains the following data:

[0065] Farmland data: Location coordinates (longitude, latitude) of each farmland, and required operation time (e.g., rotary tillage 2 hours / acre).

[0066] Agricultural machinery data: initial parking location (maintenance center) of each agricultural machine, operating efficiency, and operating cost per unit time (including fuel consumption and labor).

[0067] Route data: Based on the API of Gaode Map or Baidu Map, calculate the driving distance and estimated driving time between each pair of all location points (starting point of agricultural machinery, farmland point, and ending point of agricultural machinery) and convert them into driving cost (yuan / km).

[0068] Weather data: Obtain hourly weather forecasts for the next 72 hours from the meteorological bureau, including precipitation probability, wind speed, and temperature. Key steps: Based on agricultural machinery operation specifications (e.g., operations are not advisable when the probability of rainfall is >30%), convert the weather forecasts into feasible operation time windows for each plot of farmland. For example, if farmland F1 has no rain between 09:00 and 15:00 on the second day, its time window is set to [9, 15] (the number of hours based on 00:00 of the day).

[0069] II. Initial Optimization Model Construction (corresponding to step S1)

[0070] Based on the above data, an initial scheduling optimization model is constructed.

[0071] 1. First, the variables of the initial scheduling optimization model are designed as follows:

[0072] Farmland collection ;

[0073] Agricultural machinery collection ;

[0074] A set of location points, including farmland and start and end points. ;

[0075] : Starting point set ;

[0076] Returns the set of points. ;

[0077] Time set ;

[0078] :Location arrive Travel time;

[0079] :Location arrive The cost of driving;

[0080] Agricultural machinery farmland The unit cost of the activity;

[0081] :farmland Required operation time;

[0082] :farmland The task time window;

[0083] It can serve farmland. A collection of agricultural machinery;

[0084] Agricultural machinery The initial starting point;

[0085] Agricultural machinery Return to the destination;

[0086] : Path decision variables, if agricultural machinery From position Go to If the result is positive, then it is 1; otherwise, it is 0.

[0087] : Decision variables for job allocation, if agricultural machinery farmland If the assignment is correct, the value is 1; otherwise, it is 0.

[0088] Agricultural machinery Arrival Location Time;

[0089] 2. Objective function: Minimize the total scheduling cost. :

[0090]

[0091] Among them, activity cost Path cost .

[0092] 3. Constraint settings:

[0093] (1) Constraints on farmland operation time windows determined based on forecast weather information:

[0094] For any farmland Its operation start time Must meet ,in This refers to the feasible operation time window determined based on forecast weather information.

[0095] Each plot of farmland must begin operations within the converted time window.

[0096] (2) Continuity constraints of agricultural machinery operation paths:

[0097]

[0098]

[0099]

[0100]

[0101]

[0102] The first constraint is a path closure constraint, stipulating that each agricultural machine must depart from the designated starting point once and return to the designated ending point once to ensure the integrity of the operation path. Specifically, according to... Explanation of the determining variable Meaning: If agricultural machinery From the starting point Go to location If yes, then it is 1; otherwise, it is 0. Meaning: If agricultural machinery From position Head to the starting point Heading back to the return point If yes, then it is 1; otherwise, it is 0. The meaning is:

[0103] The second constraint is a flow balance constraint or in-degree balance constraint, which means:

[0104] For agricultural machinery and any intermediate node (Neither the beginning nor the end):

[0105] Entering the location node Path count = number of paths leaving the location node The number of paths.

[0106] Where, on the left side of the equal sign This represents all nodes that enter the intermediate position. Number of paths, decision variables The meaning is: if agricultural machinery From position Go to the middle node If the result is 1, then the result is 1; otherwise, the result is 0; the right side of the equals sign. The meaning is: if agricultural machinery From the middle position node Go to location If the result is positive, then it is 1; otherwise, it is 0.

[0107] The meaning of the third constraint: every piece of farmland It must be serviced, and can only be serviced once by one agricultural machine. To qualify to serve farmland A collection of agricultural machinery.

[0108] The meaning of the fourth constraint: If agricultural machinery To farmland Task (i.e., allocation decision) Therefore, there must be a path leading into the farmland. There must be a way to leave this farmland.

[0109] Entry constraints Decision variables The meaning is: if agricultural machinery From position Entering the farmland If yes, it is 1; otherwise, it is 0; leaving the constraint. Decision variables The meaning is: if agricultural machinery From farmland Go to location If the result is positive, then it is 1; otherwise, it is 0.

[0110] The fifth constraint is the time continuity constraint, which stipulates that if the agricultural machinery starts from a position... Go to location Then in position The start time must not be earlier than the position The sum of the start time, service time, and travel time. This constraint implicitly eliminates sub-loops in the path, ensuring the monotonically increasing nature of the time variable and its physical rationality.

[0111] For agricultural machinery Arrival Location Time;

[0112] For position Farmland operation time required;

[0113] Decision variables If agricultural machinery For position If it is farmland operation, the value is 1; otherwise, it is 0.

[0114] III. Dynamic Scheduling and Adjustment Scheme (corresponding to steps S3 to S5)

[0115] 1. Dynamic monitoring and triggering (corresponding to step S3)

[0116] In actual operation, due to weather changes, unforeseen events, or forecast errors, the originally set time windows and path feasibility may become invalid, causing some farmland to be unable to complete operations as planned. At this time, the system needs to dynamically reconstruct the scheduling plan for the remaining farmland based on the completed operations and new weather information, and generate an optimized adjustment plan.

[0117] 2. Construct a dynamic scheduling optimization model

[0118] (1) Meaning of model variables:

[0119] The collection of farmland where work has not yet been completed;

[0120] The new set of starting points composed of the current actual location of the agricultural machinery;

[0121] Agricultural machinery The current location;

[0122] : Field operation time window updated after weather changes;

[0123] (2) Objective function

[0124]

[0125] Consistent with the initial scheduling optimization model, the total scheduling cost is minimized.

[0126] (3) Model constraints

[0127] Consistent with the initial scheduling optimization model.

[0128] IV. Solving with an improved bat algorithm (corresponding to steps S2 and S5)

[0129] An improved bat algorithm is used to solve the problem. Key implementation details are as follows:

[0130] Solution Principle: This invention proposes an optimization method that nests an agricultural machinery scheduling model considering path constraints and weather time windows into the Bat Algorithm. The constructed scheduling model aims to minimize the total operating cost and path cost of agricultural machinery, while satisfying multiple constraints such as the unique access requirement of all farmland, agricultural machinery path connectivity, and weather-feasible time windows.

[0131] The Bat Algorithm, an intelligent optimization algorithm that simulates the echolocation mechanism of bats, possesses both global search capabilities and local fine-tuning capabilities, making it suitable for solving complex combinatorial optimization problems. By embedding information such as farmland allocation and access order into the particle's encoding structure, and utilizing mechanisms for frequency adjustment, velocity update, loudness, and impulse rate regulation, the Bat Algorithm can explore and evolve high-quality scheduling schemes that satisfy constraints.

[0132] This algorithm design consists of two phases: an initial optimization phase and a dynamic optimization phase. The initial phase generates a complete scheduling scheme based on predicted information; the dynamic phase performs local re-optimization based on the actual remaining tasks and new time windows when weather changes occur. Both phases share the same bat algorithm framework, differing only in data input and constraint updates.

[0133] (a) Parameter settings

[0134] : No. The solution for each particle (including field allocation and access order).

[0135] : No. The velocity of each particle is used to update its position;

[0136] : No. The frequency of each particle controls the search step size;

[0137] : No. The loudness of each particle controls the local search intensity;

[0138] : No. The pulse emission rate of each particle controls the trigger probability of local search;

[0139] : No. The fitness value of each particle measures the quality of the solution;

[0140] The globally optimal solution found during the current search process;

[0141] Maximum number of iterations

[0142] Population size (number of particles, equivalent to the set of locations).

[0143] Loudness update parameters, control Decays over time;

[0144] Pulse rate update parameters, control Increase over time;

[0145] : Penalty weight coefficient, used to constrain the penalty for violation;

[0146] Penalty for violating the time window.

[0147] Fitness function value = Total cost + Penalty term

[0148] (II) Solution Step Design

[0149] 1. Initial Scheduling Optimization Model

[0150] (1) Population initialization

[0151] First, define the key parameters of the bat algorithm, including the maximum number of iterations. Population size Frequency range and initial loudness With pulse rate Then, initialize. Each particle contains two parts: one representing farmland (…). Which agricultural machine is it assigned to? array Secondly, the order in which each agricultural machine accesses the farmland assigned to it. After initialization, assign an initial velocity to each particle. And calculate its initial fitness.

[0152] (2) Fitness assessment

[0153] For each particle First, a decoding operation is performed to restore the encoded information into a complete scheduling path and job time arrangement. Specifically, this is done according to the access order. Agricultural machinery travel time matrix Calculate the arrival time of each agricultural machine at each node in the path. .

[0154] Next, the objective value of the solution is evaluated. The total cost consists of two parts: operation cost and path cost. The operation cost is the sum of the operating costs of all allocated agricultural machinery, i.e.:

[0155]

[0156] Path cost is the sum of the costs of all agricultural machinery travel paths:

[0157]

[0158] For time window constraints, if the start time of an operation for a certain field is not within the time window, a penalty value is introduced:

[0159]

[0160] Finally, the fitness of the particles is calculated as follows:

[0161]

[0162] (3) Position and velocity update

[0163] In each iteration, for each particle, its frequency, velocity, and position are updated using the following formula:

[0164]

[0165] Position updates are achieved in the coding space through operations such as task swapping and insertion.

[0166] (4) Local search

[0167] If the random value of a certain particle This triggers a local search mechanism to construct a new solution:

[0168]

[0169] in As a disturbance factor, The search amplitude is controlled by the current loudness, and the local search is fine-tuned based on the optimal solution.

[0170] (5) Solution acceptance and parameter update

[0171] If the newly generated solution fitness If so, accept the new solution and update the loudness and impulse rate:

[0172]

[0173] (6) Update the optimal solution

[0174] After each iteration, the fitness of all particles is compared, and the current global optimum is updated. The optimal path scheme and its corresponding scheduling plan are recorded.

[0175] (7) Termination of judgment

[0176] If the maximum number of iterations is reached or the optimal solution is no longer updated for several consecutive generations, the algorithm terminates and outputs the complete scheduling scheme represented by the optimal particle, including the service objects, access order and operation start time of each agricultural machine.

[0177] 2. Dynamic scheduling optimization model

[0178] (1) Input update

[0179] Define the remaining unfinished farmland as Reacquire the operation time windows for each farmland. And record the current location of each agricultural machine as a new starting point. .

[0180] (2) Population reinitialization

[0181] Based on the remaining tasks in the optimal solution of the initial stage, a partial seed individual is constructed, and a new particle population is generated by perturbation on this basis, while keeping the original particle encoding structure unchanged.

[0182] (3) Repeat steps (2) to (7) of the initial scheduling optimization model stage.

[0183] Using the same fitness evaluation and search process as in the initial phase, a new remaining scheduling scheme is solved to ensure that unfinished tasks are completed reasonably within the new time window, and the global optimum is updated.

[0184] 3. Output and Execution

[0185] The following is an example of the dynamically optimal scheduling adjustment scheme obtained from the solution:

[0186] Agricultural Machinery M1: Depart from the maintenance center → F9 (13:00-14:00) → F10 (14:30-15:30) → Maintenance Center (Use the now-idle M1 to complete the more urgent tasks originally assigned to M3).

[0187] Agricultural Machinery M2: From the current location on the way, return directly to the repair center to avoid the weather (cancel the original F7 task).

[0188] Agricultural Machinery M3: After completing F6 operation → return to the maintenance center → depart again at 18:00 → F7 (18:30-20:30) → F8 (21:00-23:00) → maintenance center (F7 and F8 will be postponed to safe time periods).

[0189] The system immediately pushes the adjusted path and time window to the relevant agricultural machinery operator terminals to guide them in executing the new plan.

[0190] While the invention has been described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the invention. Therefore, it should be understood that many modifications can be made to the exemplary embodiments, and other arrangements can be designed without departing from the spirit and scope of the invention as defined by the appended claims. It should be understood that different dependent claims and features described herein can be combined in ways different from those described in the original claims. It is also understood that features described in conjunction with individual embodiments can be used in other described embodiments.

Claims

1. A method for adjusting agricultural machinery operation plans considering dynamic weather changes, characterized in that, Includes the following steps: S1. Construct an initial scheduling optimization model whose objective function is to minimize the total scheduling cost, with the following constraints: Constraints on farmland operation time windows determined based on forecast weather information; Continuity constraints on agricultural machinery operation paths; S2. The improved bat heuristic algorithm is used to solve the initial scheduling optimization model and output the initial optimal scheduling scheme, which includes the operation path, operation sequence and operation time arrangement of each agricultural machine. S3. Real-time monitoring of weather conditions. When actual weather changes are detected that cause the time window for farmland operations to fail or the driving path of agricultural machinery to become infeasible, a dynamic adjustment mechanism is triggered, and the following dynamic adjustment steps are initiated. S4. Based on the current completed task status, real-time location information of agricultural machinery, and updated weather information, a dynamic scheduling optimization model is constructed. Its objective function is to minimize the total scheduling cost of the remaining tasks. Constraints include: Constraints on farmland operation time windows redefined based on updated weather information; Path continuity constraints for agricultural machinery starting from its current position; S5. The improved bat heuristic algorithm is used to solve the dynamic scheduling optimization model, and the dynamic optimal scheduling scheme is output to guide the execution of the remaining tasks.

2. The method for adjusting agricultural machinery operation plans considering dynamic weather changes according to claim 1, characterized in that, In step S1, the objective function of the initial optimization model is: In the formula, For the total scheduling cost, , For operating costs, For path cost; For agricultural machinery farmland The unit cost of the operation. , For agricultural machinery collection, , Collect farmland; For agricultural machinery from position arrive The cost of driving , , For the set of location points; As path decision variables, if agricultural machinery From position Go to If yes, then it is 1; otherwise, it is 0. Assign decision variables to the operation: if agricultural machinery farmland If the assignment is correct, the value is 1; otherwise, it is 0.

3. The method for adjusting agricultural machinery operation plans considering dynamic weather changes according to claim 1, characterized in that, In step S1, the operation time window constraint is: for any farmland Its operation start time Must meet ,in This refers to the feasible operation time window determined based on forecast weather information.

4. The method for adjusting agricultural machinery operation plans considering dynamic weather changes according to claim 3, characterized in that, In steps S2 and S5, the improved bat heuristic algorithm adds the following mechanism to the standard bat algorithm: Coding Design: The scheduling scheme is encoded into a farm machinery-farmland allocation vector and a farm machinery internal access order vector; Fitness function design: fitness value The expression is: , The penalty value for violating the time window; Local search strategy: When the local search triggering condition is met, a new solution is constructed by introducing random perturbation based on the current optimal solution.

5. The method for adjusting agricultural machinery operation plans considering dynamic weather changes according to claim 1, characterized in that, In step S4, the dynamic scheduling optimization model is adjusted as follows based on the initial scheduling optimization model: the set of farmland, the set of current locations of agricultural machinery, and the set of operation time windows are updated.

6. The method for adjusting agricultural machinery operation plans considering dynamic weather changes according to claim 5, characterized in that, The solution process of the dynamic scheduling optimization model includes: Based on the portion of the initial optimal scheduling scheme output in step S2 concerning the remaining tasks, construct a partially retained seed individual; Perturbation is performed on the seed individuals to generate a new optimized population; The new population is optimized using the same bat algorithm framework as in step S2.

7. The method for adjusting agricultural machinery operation plans considering dynamic weather changes according to claim 1, characterized in that, The method also includes outputting the operation path, operation sequence, and time schedule for each agricultural machine.

Images