A task scheduling method and device for smart workshop

Abstract

The present invention discloses a task scheduling method and device for a smart workshop, which relates to the technical field of task scheduling. The power consumption data of a target production line is obtained to calculate the power fluctuation abnormal value of the target production line; the production timestamp of the product on the target production line is obtained to calculate the production time deviation of the target production line, and the failure probability of the target production line is determined by combining the production time deviation and the power fluctuation abnormal value; if the failure probability is greater than a preset failure threshold, the target production line is recorded as the first production line, otherwise, the target production line is recorded as the second production line; the currently remaining production tasks of each first production line in the workshop are allocated to each second production line; in this way, production line failures can be discovered in time and bad production tasks can be stopped, the impact of the failure can be reduced, and product quality can be ensured; at the same time, unfinished production tasks will be reallocated to other normal production lines to ensure production speed and minimize overall failures, thereby improving the efficiency of workshop production task scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of task scheduling, and in particular to a task scheduling method and device for a smart workshop. Background Art

[0002] In modern manufacturing workshops, production lines are the core facilities for achieving production tasks. In order to improve production efficiency and respond to market demand, workshops usually choose to operate multiple production lines in parallel to jointly complete a batch of production tasks, so as to maximize production capacity and optimize resource allocation. In particular, for production tasks that need to be delivered quickly, by operating multiple production lines in parallel, workshops can improve production efficiency and reduce downtime waiting time.

[0003] However, during the production process, some production lines may malfunction or have other problems. Continuing to execute production tasks may aggravate the production line failures and may even affect product quality. In order to avoid product quality problems and delays in production time, the affected production lines must be suspended and their unfinished production tasks must be reallocated to other normally operating production lines. However, blindly allocating tasks may lead to unreasonable production arrangements, slower production speeds, and may even further increase the failures of certain production lines, resulting in poor scheduling of workshop production tasks. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and provide a task scheduling method and device for a smart workshop.

[0005] In a first aspect of the implementation of the present invention, a task scheduling method for a smart workshop is first proposed, the method comprising:

[0006] Acquire power consumption data of a target production line, and calculate a power fluctuation abnormality value of the target production line according to the power consumption data; the target production line is any one of a plurality of production lines in a workshop;

[0007] Obtaining a production timestamp of the product on the target production line, calculating a production time deviation of the target production line according to the production timestamp, and determining a failure probability of the target production line by combining the production time deviation with the power fluctuation abnormal value;

[0008] If the failure probability is greater than a preset failure threshold, the target production line is recorded as the first production line, otherwise, the target production line is recorded as the second production line;

[0009] Allocate the remaining production tasks of each first production line in the workshop to each second production line.

[0010] Optionally, the step of calculating the power fluctuation abnormal value of the target production line according to the power consumption data is:

[0011] Each production equipment of the target production line is recorded as a target production equipment, and the time interval from the start of the production task of the target production equipment to the current time is recorded as a target time interval, and the power of the target production equipment at each moment in the target interval is obtained to obtain the power signal of the target production equipment;

[0012] Apply Fourier transform to the target production equipment power signal to convert the time domain signal into a frequency domain signal;

[0013] The part of the frequency domain signal that is higher than the preset frequency threshold is regarded as the high-frequency area, and the part that is not higher than the preset frequency threshold is regarded as the low-frequency area, and the total energy of the high-frequency area is calculated respectively. and the total energy in the low-frequency region ;

[0014] according to and Calculate the power fluctuation anomaly ratio of the target production equipment , ;

[0015] Calculate the average power fluctuation anomaly ratio of all target production equipment to obtain the power fluctuation anomaly value of the target production line.

[0016] Optionally, the step of calculating the production time deviation of the target production line according to the production timestamp is:

[0017] The time from the start of the production task to the current time of the target production line is recorded as the preset time interval, and the timestamp of the preset production completion of each product on the target production line within the preset time interval is obtained to obtain the preset production time sequence based on the time order. , ;According to the timestamp of each product's actual production completion, the actual production time series based on time order is obtained , , and each time series has numerical values, Indicates the total number of products produced;

[0018] Defining the distance function , ;in and Respectively represent The timestamp of the preset production completion and the timestamp of the actual production completion of each product;

[0019] Construct a The cumulative distance matrix ,in From the first time point to the and The cumulative minimum distance of time points satisfies the following recursive relationship:

[0020] ;

[0021] And the initial conditions ; ; ;

[0022] final From the sequence arrive The shortest path of the target production line; calculate the production time deviation of the target production line , the calculation formula is: .

[0023] Optionally, the step of determining the failure probability of the target production line in combination with the production time deviation and the power fluctuation abnormal value is:

[0024] The power fluctuation abnormal values ​​and production time deviations of the products on each target production line are normalized, and the failure probability of the corresponding target production line is calculated based on the normalized power fluctuation abnormal values ​​and production time deviations. The calculation expression is:

[0025] ;

[0026] In the formula, is the failure probability of the target production line, , are the normalized power fluctuation anomaly and production time deviation, Respectively , The preset scaling factor of Both are greater than 0.

[0027] Optionally, if the failure probability is greater than a preset failure threshold, the target production line is recorded as the first production line, otherwise, the target production line is recorded as the second production line; when the target production line is the second production line, the target production line immediately sends an alarm signal.

[0028] Optionally, the steps of allocating the currently remaining production tasks of each first production line in the workshop to each second production line are:

[0029] Obtain the current quantity of products to be produced by each second production line and the remaining quantity of all products to be produced by each first production line, and sum them up to obtain the remaining quantity of products to be produced;

[0030] The production rate, failure probability, remaining maximum number of producible products and maximum power limit of each second production line are obtained, and the remaining number of products to be produced is reallocated to each second production line to obtain the production task of each second production line.

[0031] Optionally, the remaining number of products to be produced is reallocated to each second production line, and the specific steps of obtaining the production task of each second production line are:

[0032] Define the objective function: minimize the total production time and minimize the probability of failure;

[0033] Define constraints: the task volume assigned to each second production line does not exceed its remaining maximum number of products that can be produced, the power consumption of each second production line does not exceed its maximum power limit, and the remaining number of products to be produced must be fully allocated to each second production line;

[0034] Inputting the objective function and the constraint conditions into the mixed integer linear programming solver to solve the objective function and the constraint conditions;

[0035] The optimal solution obtained by the mixed integer linear programming solver is used to obtain the number of products to be produced on each second production line, which serves as the production task of each second production line.

[0036] In a second aspect of the present invention, a task scheduling device for a smart workshop is provided, the device comprising:

[0037] Power fluctuation abnormality module: for each production line currently executing a production task, record it as a target production line, obtain the power data consumed by each production equipment on each target production line during operation, and calculate the power fluctuation abnormality value of the corresponding production line;

[0038] Fault module: obtains the production timestamps of the products on each target production line, calculates the production time deviation of the production line, and combines the power fluctuation abnormality value to obtain the fault probability of the corresponding target production line;

[0039] Screening module: Screen the target production line according to the failure probability and the preset failure probability threshold, and the production line that can continue to produce products is recorded as the second production line;

[0040] Task scheduling module: obtains the current product quantity to be produced, production rate and failure probability of each second production line, and obtains all remaining product quantities to be produced by the production lines that can no longer produce products, and distributes the second product quantities to the production tasks of each second production line.

[0041] Beneficial effects of the present invention:

[0042] 1. The present invention proposes a task scheduling method and device for a smart workshop, which obtains the power consumption data of a target production line, calculates the power fluctuation abnormality of the target production line according to the power consumption data; obtains the production timestamp of the product on the target production line, calculates the production time deviation of the target production line according to the production timestamp, and determines the failure probability of the target production line by combining the production time deviation and the power fluctuation abnormality; if the failure probability is greater than a preset failure threshold, the target production line is recorded as the first production line, otherwise, the target production line is recorded as the second production line; this process can effectively ensure the safety and efficiency of production, while maximally reducing the possible risks and losses in the production process, reducing the failure of the production line, and ensuring that it will not affect the product quality;

[0043] 2. By allocating the remaining production tasks of each first production line in the workshop to each second production line, the unfinished production tasks on the affected production lines can be reasonably reallocated to other normally operating production lines, ensuring the production speed while minimizing the overall production line failures, thereby achieving better workshop production task scheduling effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] Figure 1 is a flow chart of a task scheduling method for a smart workshop;

[0046] Figure 2 The framework diagram of a task scheduling device for a smart workshop. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0049] The embodiment of the present invention provides a task scheduling method for a smart workshop. Figure 1 , Figure 1 A flowchart of a task scheduling method for a smart workshop provided by an embodiment of the present invention. The method comprises the following steps:

[0050] Acquire power consumption data of a target production line, and calculate a power fluctuation abnormality value of the target production line according to the power consumption data; the target production line is any one of multiple production lines in a workshop;

[0051] Obtain the production timestamp of the product on the target production line, calculate the production time deviation of the target production line based on the production timestamp, and determine the failure probability of the target production line by combining the production time deviation and the power fluctuation abnormal value;

[0052] If the failure probability is greater than the preset failure threshold, the target production line is recorded as the first production line, otherwise, the target production line is recorded as the second production line;

[0053] Allocate the remaining production tasks of each first production line in the workshop to each second production line.

[0054] Based on a task scheduling method for a smart workshop provided by an embodiment of the present invention, through the above-mentioned method, if some production lines may have failures or other problems, they can be discovered in time and the execution of production tasks can be stopped, thereby reducing the failure of the production line and ensuring that the product quality will not be affected; and the unfinished production tasks on the affected production lines can be reasonably reallocated to other normally operating production lines, thereby ensuring the production speed while minimizing the overall production line failures, thereby achieving a better workshop production task scheduling effect.

[0055] In one embodiment, the steps of calculating the power fluctuation abnormality value of the target production line according to the power consumption data are as follows:

[0056] Each production equipment of the target production line is recorded as the target production equipment, and the time interval from the start of the production task to the current time of the target production equipment is recorded as the target time interval. The power of the target production equipment at each moment in the target interval is obtained to obtain the power signal of the target production equipment; recorded as , Indicates the total number of power signals;

[0057] The target production equipment power signal As a time domain signal, Apply Fourier transform to transform the time domain signal Convert to frequency domain signal , Indicates Frequency components ;

[0058] Calculate the magnitude of each frequency component , the calculation formula is: ;in, is a frequency domain signal The module is: ;in, and are the real and imaginary parts of the Fourier transform result respectively;

[0059] The frequency domain signal above the preset frequency threshold The part is regarded as the high-frequency area, and the total energy of the high-frequency area is calculated , the calculation formula is: ;in, It is the square of the spectrum amplitude, representing the energy of each frequency component, and calculating the total energy in the low-frequency area. Calculated as: ; Calculate the power fluctuation anomaly ratio of the target production equipment based on the total energy in the high-frequency area and the energy in the low-frequency area , ;

[0060] Calculate the average power fluctuation anomaly ratio of all target production equipment to obtain the power fluctuation anomaly value of the target production line.

[0061] It should be noted that the preset frequency threshold It is set by professionals based on actual conditions and is not limited or elaborated on in detail.

[0062] It should be noted that when calculating the power fluctuation anomaly value of the target production line, the target production line can still produce items, so there will not be a phenomenon that the normal operation of the entire production line will be affected due to a large abnormality of a certain production equipment. At this time, in order to determine whether the target production line has a fault, continuing to execute the production task may aggravate the fault of the production line and may even affect the product quality. The power fluctuation anomaly ratio of each production equipment on the target production line is calculated to determine the target production line fault; and the average power fluctuation anomaly ratio of all production equipment is calculated to continue to analyze and determine whether the target production line has a fault.

[0063] It should be noted that the power consumed at different times when the target production equipment is running refers to the power consumption of the equipment that changes over time during its operation. The power consumption of different target production equipment will be affected by multiple factors when working, such as workload, operating speed, equipment status, etc. Power fluctuations can often reflect abnormal conditions in equipment operation, such as uneven load, increased friction, wear or failure of parts, etc. These fluctuations usually become more significant when the equipment fails or operates abnormally. Therefore, by monitoring the power changes of the equipment, potential problems can be discovered in time. Power fluctuations are a direct indicator of the working status of the equipment and can accurately reflect the operating health of the equipment. A sudden increase or decrease in power may mean that the equipment is overloaded or malfunctioning, while continuous abnormal fluctuations may indicate that the equipment has systemic problems or is about to fail. The power data of each production equipment can usually be obtained in real time through sensors in the industrial automation system, such as power meters, which can accurately monitor the power consumption of the equipment under different working conditions, thereby providing reliable data support for subsequent analysis.

[0064] It should be noted that the time when the target production equipment starts the production task and the current time can be obtained through the monitoring log of the production line, or other acquisition methods can be used, which are not limited or elaborated in detail.

[0065] It should be noted that the larger the abnormal value of the power fluctuation of the production line, the greater the possibility of failure of the corresponding target production line. Continuing to perform the production task may aggravate the failure of the production line and may even affect the product quality. The reason is that the larger the abnormal value of the power fluctuation of the production line, the more severe the load and power fluctuation of the equipment, which is often a signal of unstable equipment state or potential failure; when the equipment runs for a long time under abnormal load fluctuation, it may increase the wear of the equipment, increase the fatigue of mechanical parts, and even cause failure. At the same time, uneven load of equipment in the production process may lead to unstable production quality of products, such as dimensional accuracy, surface quality, etc., which do not meet the standards, and may even cause batch defects; in addition, abnormal equipment load may also cause changes in other abnormal parameters such as temperature and vibration, thereby affecting the stability and accuracy of the entire production process; if these abnormal production lines continue to perform tasks, it will not only increase the risk of equipment failure, but also may cause more waste, rework and even customer complaints due to substandard product quality, which will ultimately affect the production efficiency and market reputation of the enterprise. Therefore, timely identification and handling of abnormal power fluctuations in production lines is the key to ensuring production efficiency and product quality.

[0066] In one implementation, the benefits of calculating the power fluctuation anomaly of the production line in the above manner are: using Fourier transform to convert the time domain signal into a frequency domain signal can reveal the periodic and non-periodic components of the power fluctuation of the target production equipment, especially the energy changes in the high-frequency region, which are usually closely related to equipment wear, overload or impending failure; by setting a preset frequency threshold and distinguishing between high-frequency and low-frequency regions, the type of power fluctuation can be analyzed more specifically, and normal fluctuations can be distinguished from abnormal fluctuations. In general, the energy in the low-frequency region corresponds to periodic fluctuations related to the normal operation of the production equipment (such as normal vibration of the motor, load changes, etc.); while the fluctuations in the high-frequency region It usually reflects the abnormal fluctuation of equipment more directly, which is often related to equipment failure. By calculating the energy proportion in the high-frequency area, it can avoid interference with normal operation due to misjudgment. In addition, this method can timely discover the potential failure risk of the target production equipment, avoid continuing to execute production tasks when the target production equipment is in poor condition, reduce the impact of the target production equipment failure on the target production line, and ensure the stability of product quality. By calculating the power fluctuation abnormality ratio of all target production equipment and taking the average, it can provide an overall assessment of the abnormal load level of the target production line, so that the health status of the production line can be comprehensively and objectively monitored, thereby providing data support for subsequent production scheduling and maintenance decisions.

[0067] In one embodiment, the steps of calculating the production time deviation of the target production line according to the production timestamp are:

[0068] The time from the start of the production task to the current time of the target production line is recorded as the preset time interval, and the timestamp of the preset production completion of each product on the target production line within the preset time interval is obtained to obtain the preset production time sequence based on the time order. , ;According to the timestamp of each product's actual production completion, the actual production time series based on time order is obtained , , and each time series has numerical values, Indicates the total number of products produced;

[0069] Defining the distance function , usually the absolute difference is used as the local distance, the expression is: ;in, and Respectively represent The timestamp of the preset production completion and the timestamp of the actual production completion of each product;

[0070] In order to calculate the minimum distance between two time series, the DTW algorithm is used to construct a cumulative distance matrix; specifically: construct a The cumulative distance matrix ,in From the first time point to the and The cumulative minimum distance of time points satisfies the following recursive relationship:

[0071] ,

[0072] And the initial conditions ; ; ;

[0073] The minimum distance of DTW is from the matrix The last element of is obtained, that is: , this value represents the "shortest path" between the actual production time series and the preset production time series; that is, the similarity measure between the two time series; the smaller this distance is, the smaller the difference between the two time series is;

[0074] final From the sequence arrive The shortest path of the target production line; calculate the production time deviation of the target production line , the calculation formula is: .

[0075] It should be noted that the preset production time series needs to be set by professionals and generate preset completion timestamps for each product based on production plans, task requirements and specific production lines, combined with parameters such as preset time intervals and production rates; in addition, the actual production completion timestamp of each product can be obtained through sensors, automation equipment or production management systems on the production line. When the production equipment completes the production or processing operation of the product, the system will automatically record the production completion time of the product and associate the timestamp with the unique identifier of the product (such as a barcode or RFID tag). It can also be obtained in other ways, which are not limited or elaborated on.

[0076] It should be noted that the production time deviation of a production line refers to the degree of deviation between the actual completion time of the product produced by the target production line from the time when the current production task is started to the current time and the preset completion time; if the degree of deviation is larger, it means that the possibility of failure of the corresponding target production line is greater, and continuing to execute the production task may aggravate the failure of the production line and may even affect the quality of the product. The reason is that if the degree of deviation is too large, it means that the equipment or process on the production line may have abnormalities or reduced efficiency, resulting in the production process not being completed smoothly according to the scheduled time node; this deviation may be caused by equipment failure, improper operation of personnel, raw material problems or other external factors; continuing to execute the production task may aggravate these potential problems, especially when the production line has already experienced a certain degree of failure, continuing to operate will not only delay the production progress, but also may have a negative impact on the quality of the product; for example, abnormalities in production equipment may lead to reduced processing accuracy, or unqualified products during the production process, which will affect the quality of the entire batch of products and increase the scrap rate and rework rate; therefore, stopping the affected production line in time and troubleshooting can effectively avoid quality problems and production delays, thereby ensuring the stability of production and the quality of the product.

[0077] In one implementation, the benefits of using the above method to calculate the production time deviation of the production line are: the DTW algorithm can handle nonlinear changes between time series, especially in the production process, the actual production time is often affected by multiple factors such as equipment status, process flow, and personnel operation, which may cause fluctuations in production time; unlike the traditional simple time difference calculation method, DTW can more comprehensively evaluate the deviation in the production process by comparing the cumulative differences of the entire time series, avoiding the impact of local errors at a single time point on the overall evaluation. In addition, DTW can dynamically adjust the comparison method to adapt to time fluctuations in the production process, providing more accurate fault prediction and problem location; ultimately, this method can effectively improve the accuracy of production line fault warnings, timely discover potential risks and take targeted measures, thereby reducing production delays, improving product quality and optimizing resource allocation.

[0078] In one embodiment, the steps of determining the failure probability of the target production line by combining the production time deviation and the power fluctuation abnormal value are as follows:

[0079] The power fluctuation abnormal values ​​and production time deviations of the products on each target production line are normalized, and the failure probability of the corresponding target production line is calculated based on the normalized power fluctuation abnormal values ​​and production time deviations. The calculation expression is:

[0080] ;

[0081] In the formula, is the failure probability of the target production line, , are the normalized power fluctuation anomaly and production time deviation, Respectively , The preset scaling factor of Both are greater than 0.

[0082] It should be noted that Set according to actual situation. The sum is 1, for example, The value of can be 0.5, 0.5, or other values, which depends on the actual situation and is not limited or elaborated on. In addition, commonly used normalization processing methods include Min-Max normalization, Z-Score standardization, etc., which depends on the actual situation and is not limited or elaborated on.

[0083] In one embodiment, the failure probability of the target production line is compared with a preset failure probability threshold. If the failure probability is not less than the preset failure probability threshold, the target production line cannot continue to produce products, the target production line is recorded as the first production line, and the production work of the target production line is immediately stopped, and an alarm signal is issued.

[0084] If the failure probability is less than the preset failure probability threshold, the target production line can continue to produce products, and the target production line is recorded as the second production line;

[0085] It should be noted that the preset fault probability threshold is set by professionals based on actual conditions and will not be limited or elaborated on in detail.

[0086] It should be noted that the process of screening the target production lines according to the failure probability and the preset failure probability threshold is aimed at ensuring the stability of the production line and the product quality; in this process, the higher the failure probability of each target production line, it means that the target production line may have a greater failure risk or production instability; these failure probabilities are compared with the preset failure probability threshold; if the failure probability of a target production line is less than the preset threshold, it indicates that the status of the production line is still within an acceptable range and can continue to perform production tasks; at this time, this production line will be regarded as the "second production line" and continue to participate in the current production tasks.

[0087] However, if the failure probability of the target production line is not less than the preset failure probability threshold, it means that the corresponding target production line has a high risk of failure, and continued production may cause more problems or affect product quality; in this case, in order to prevent potential quality problems or production accidents, the production of the target production line must be stopped immediately; at this time, the target production line will automatically send out an alarm signal to notify relevant personnel to conduct inspections and maintenance to ensure that the problem is handled in a timely manner; the alarm signal can also trigger the automatic adjustment of the production scheduling system, reallocating the tasks originally undertaken by the target production line to other normally operating production lines, thereby avoiding production delays or product quality degradation. Through such a screening and processing mechanism, the safety and efficiency of production can be effectively guaranteed, while minimizing possible risks and losses in the production process.

[0088] In one embodiment, the steps of allocating the remaining production tasks of each first production line in the workshop to each second production line are:

[0089] Obtain the current quantity of products to be produced by each second production line and the remaining quantity of all products to be produced by each first production line, and sum them up to obtain the remaining quantity of products to be produced;

[0090] The production rate, failure probability, remaining maximum number of producible products and maximum power limit of each second production line are obtained, and the remaining number of products to be produced is reallocated to each second production line to obtain the production task of each second production line.

[0091] It should be noted that the current amount of products to be produced by each second production line and the remaining amount of all products to be produced by each first production line can be obtained through the real-time task scheduling system or production progress tracking system of the production line, and the data is usually recorded through the production plan or on-site material tracking system; the production rate can be obtained through the monitoring system of the production line, and the rate data is usually calculated by equipment sensors and production control systems; the remaining maximum number of producible products can be obtained from the production planning system or inventory management system, which is estimated by professionals based on the current production task volume, the production capacity of the equipment and the remaining available time; and the maximum power limit is set by professionals according to the actual situation; in addition, the power consumed per unit product needs to be obtained through the real-time energy consumption monitoring system of the production line; it can also be other acquisition methods, which are not limited or elaborated on.

[0092] In one embodiment, the remaining number of products to be produced is reallocated to each second production line to obtain the production task of each second production line. The specific steps are:

[0093] S1: Define the objective function:

[0094] S1.1: Minimize the total production time (i.e., minimize the total time to complete the production of the remaining number of products to be produced); the specific formula is: , where Indicates that the remaining number of products to be produced is allocated to the second production line The number of products on Indicates the total number of the second production line; Indicates the second production line production rate;

[0095] S1.2: Minimize the failure probability (i.e., minimize the failure probability corresponding to the number of remaining products to be produced to complete production); the specific formula is: , where Indicates the second production line The probability of failure;

[0096] S2: Define constraints:

[0097] S2.1: The task volume assigned to each second production line does not exceed the maximum number of products that can be produced, that is: , ,in, The second production line The remaining maximum number of products that can be produced;

[0098] S2.2: The power consumption of each second production line does not exceed its maximum power limit, that is: , ,in, The second production line The power consumed per unit of product; For the second production line Maximum power limit;

[0099] S2.3: The remaining number of products to be produced must be fully allocated to each second production line, that is, , where is the number of products remaining to be produced;

[0100] S3: Input the objective function and constraints into the mixed integer linear programming solver to solve the following optimization problem: ; ; ; ; ;

[0101] S4: Obtain the optimal solution obtained by the mixed integer linear programming solver to obtain the number of products to be produced on each second production line.

[0102] It should be noted that the mixed integer linear programming solver (MILP Solver) is an optimization algorithm specifically used to solve linear programming problems with integer variables. In mixed integer linear programming problems, decision variables may not only be continuous (such as real numbers) but also integers (such as the production quantity of a product), and the solver needs to be able to handle mixed variables; MILP solvers usually contain the following parts:

[0103] Objective Function: This is the function that needs to be optimized, usually a quantity to be maximized or minimized, such as cost, time, revenue, etc. In the production task allocation problem, the objective function may be to minimize production time or failure probability.

[0104] Constraints: Constraints limit the range of variables to ensure the feasibility of the solution. For example, the maximum power of the production line, the production capacity of each production line, the number of remaining products, etc. are all constraints.

[0105] Decision Variables: These variables represent the quantity to be optimized, such as the number of products to be assigned to each production line. They may be integer (such as the number of products produced) or continuous (such as production time).

[0106] Solving Algorithm: The MILP solver uses various optimization algorithms (such as the simplex method, branch and bound method, cutting plane method, etc.) to explore possible values ​​of the decision variables and find the optimal solution based on the objective function.

[0107] In the scenario of production task allocation, by inputting the objective function (such as minimizing production time or failure probability) and constraints (such as the remaining maximum number of producible products, power limit, remaining number of products, etc.) into the MILP solver, the solver will calculate according to these conditions and optimize the decision variables through an iterative process, and finally obtain the number of production products that should be allocated to each production line. In specific implementation, the solver will select the optimal solution from a set of possible solutions, that is, the solution that minimizes the objective function and satisfies all constraints. This method can effectively balance the task load between different production lines and ensure production efficiency and safety. Through the MILP solver, the most suitable product allocation solution can be found under complex constraints.

[0108] In one implementation method, through the optimization calculation of the mixed integer linear programming (MILP) solver, the load can be balanced among multiple production lines to avoid excessive load or idleness of a certain production line, thereby improving the overall production capacity; secondly, reasonable task allocation can reduce resource waste in the production process, such as reducing downtime, excessive power consumption and other problems, improving the stability of the production line and extending the service life of the equipment; in addition, optimizing production tasks can also minimize the probability of failure and production delays, reduce the failure rate and improve product quality under limited resources (such as power and maximum production capacity).

Claims

1. A task scheduling method for a smart workshop, characterized in that: The following steps are involved: Acquire power consumption data of a target production line, and calculate a power fluctuation abnormality value of the target production line according to the power consumption data; the target production line is any one of a plurality of production lines in a workshop; Obtaining a production timestamp of the product on the target production line, calculating a production time deviation of the target production line according to the production timestamp, and determining a failure probability of the target production line by combining the production time deviation with the power fluctuation abnormal value; If the failure probability is greater than a preset failure threshold, the target production line is recorded as the first production line, otherwise, the target production line is recorded as the second production line; Allocate the remaining production tasks of each first production line in the workshop to each second production line; The steps of calculating the production time deviation of the target production line according to the production timestamp are: The time from the start of the production task to the current time of the target production line is recorded as the preset time interval, and the timestamp of the preset production completion of each product on the target production line within the preset time interval is obtained to obtain the preset production time sequence based on the time order. , ;According to the timestamp of each product's actual production completion, the actual production time series based on time order is obtained , , and each time series has numerical values, Indicates the total number of products produced; Defining the distance function , ;in and Respectively represent The timestamp of the preset production completion and the timestamp of the actual production completion of each product; Construct a The cumulative distance matrix , From the first time point to the and The cumulative minimum distance of time points satisfies the following recursive relationship: ； And the initial conditions ; ; ; final From the sequence arrive The shortest path of the target production line; calculate the production time deviation of the target production line , the calculation formula is: .

2. A task scheduling method for a smart workshop according to claim 1, characterized in that: The step of calculating the power fluctuation abnormal value of the target production line according to the power consumption data is as follows: Each production equipment of the target production line is recorded as a target production equipment, and the time interval from the start of the production task of the target production equipment to the current time is recorded as a target time interval, and the power of the target production equipment at each moment in the target interval is obtained to obtain the power signal of the target production equipment; Applying Fourier transform to the power signal of the target production equipment to convert the time domain signal into a frequency domain signal; The part of the frequency domain signal that is higher than the preset frequency threshold is regarded as the high-frequency area, and the part that is not higher than the preset frequency threshold is regarded as the low-frequency area, and the total energy of the high-frequency area is calculated respectively. and the total energy in the low-frequency region ; according to and Calculate the power fluctuation anomaly ratio of the target production equipment , ; Calculate the average power fluctuation anomaly ratio of all target production equipment to obtain the power fluctuation anomaly value of the target production line.

3. The task scheduling method for a smart workshop according to claim 1 is characterized in that: The step of determining the failure probability of the target production line by combining the production time deviation and the power fluctuation abnormal value is as follows: The power fluctuation abnormal values ​​and production time deviations of the products on each target production line are normalized, and the failure probability of the corresponding target production line is calculated based on the normalized power fluctuation abnormal values ​​and production time deviations. The calculation expression is: ； In the formula, is the failure probability of the target production line, , are the normalized power fluctuation anomaly and production time deviation, Respectively , The preset scaling factor of Both are greater than 0.

4. The task scheduling method for a smart workshop according to claim 1 is characterized in that: When the target production line is the second production line, the target production line immediately sends out an alarm signal.

5. The task scheduling method for a smart workshop according to claim 1 is characterized in that: The steps to allocate the remaining production tasks of each first production line in the workshop to each second production line are: Obtain the current quantity of products to be produced by each second production line and the remaining quantity of all products to be produced by each first production line, and sum them up to obtain the remaining quantity of products to be produced; The production rate, failure probability, remaining maximum number of producible products and maximum power limit of each second production line are obtained, and the remaining number of products to be produced is reallocated to each second production line to obtain the production task of each second production line.

6. A task scheduling method for a smart workshop according to claim 5, characterized in that: The remaining number of products to be produced is reallocated to each second production line, and the specific steps of obtaining the production task of each second production line are: Define the objective function: minimize the total production time and minimize the probability of failure; Define constraints: the task volume assigned to each second production line does not exceed its remaining maximum number of products that can be produced, the power consumption of each second production line does not exceed its maximum power limit, and the remaining number of products to be produced must be fully allocated to each second production line; Inputting the objective function and the constraint conditions into the mixed integer linear programming solver to solve the objective function and the constraint conditions; The optimal solution obtained by the mixed integer linear programming solver is used to obtain the number of products to be produced on each second production line, which serves as the production task of each second production line.

7. A task scheduling device for a smart workshop, used to implement a task scheduling method for a smart workshop as described in any one of claims 1 to 6, characterized in that: The device comprises: Power fluctuation abnormality module: obtains power consumption data of a target production line, and calculates a power fluctuation abnormality value of the target production line according to the power consumption data; the target production line is any one of multiple production lines in a workshop; Fault probability module: obtains the production timestamp of the product on the target production line, calculates the production time deviation of the target production line according to the production timestamp, and determines the fault probability of the target production line by combining the production time deviation with the power fluctuation abnormal value; Judgment module: if the failure probability is greater than a preset failure threshold, the target production line is recorded as the first production line; otherwise, the target production line is recorded as the second production line; Task scheduling module: allocates the remaining production tasks of each first production line in the workshop to each second production line.

Images