A multi-objective collaborative scheduling optimization method and system for an intelligent manufacturing workshop
By dynamically adjusting the weights of workshop production parameters, the shortcomings of the fixed-weight method in scheduling optimization when facing order fluctuations and sudden changes in energy consumption are solved, and more efficient multi-objective collaborative scheduling is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU HEMA INFORMATION TECH CO LTD
- Filing Date
- 2025-08-08
- Publication Date
- 2026-05-26
AI Technical Summary
In existing technologies, the fixed-weight linear summation method cannot adjust the weights in a timely manner when faced with order fluctuations or sudden changes in production energy consumption prices, resulting in poor optimization of workshop scheduling.
By obtaining the scheduling execution results within a preset time period, the initial weights are adjusted based on the production cost variable, and the correction coefficient is determined by combining the degree of conflict within the weight adjustment period. The weights are then dynamically adjusted to adapt to the industrial production environment.
It enables timely adjustment of weights in a dynamic environment, improving the accuracy and effectiveness of multi-objective scheduling optimization in the workshop.
Smart Images
Figure CN121010143B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data computing technology, and in particular to a multi-objective collaborative scheduling optimization method and system for intelligent manufacturing workshops. Background Technology
[0002] With the in-depth advancement of the intelligent manufacturing strategy, multi-objective collaborative scheduling has become a core requirement of leading factories. The core purpose of workshop multi-objective collaborative scheduling optimization is to optimize multiple conflicting or related objectives simultaneously in a complex manufacturing environment by systematically coordinating scheduling decisions of multiple links and resources, thereby improving overall production efficiency and competitiveness.
[0003] Currently, the common method for multi-objective collaborative scheduling optimization in workshops is the fixed-weight linear summation method, which sums the various objective indicators with corresponding weights. Although this method can optimize multi-objective scheduling, these weights rely heavily on expert experience or static data and cannot adapt to the dynamic environment of industrial production. When faced with order fluctuations or sudden changes in production energy consumption prices, the weights cannot be adjusted in a timely manner, resulting in poor scheduling optimization effects in the workshop. Summary of the Invention
[0004] The main purpose of this application is to provide a multi-objective collaborative scheduling optimization method and system for intelligent manufacturing workshops. It aims to solve the technical problem in related technologies where a fixed-weight linear summation method is used to sum various objective indicators with corresponding weights. However, when faced with order fluctuations or sudden changes in production energy consumption prices, the weights cannot be adjusted in a timely manner, resulting in poor scheduling optimization effects in the workshop.
[0005] To achieve the above objectives, embodiments of this application provide a multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops, the method comprising:
[0006] Obtain the scheduling execution results obtained by scheduling workshop production parameters according to initial weights within a preset time period;
[0007] Based on the scheduling execution results and production cost variables, the initial weights of different dimensions are adjusted to obtain the adjusted weights;
[0008] The correction coefficient is determined based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period;
[0009] The adjusted weights are corrected based on the correction factor to obtain the corrected weights;
[0010] The workshop production parameters are scheduled and optimized using the corrected weights to obtain the scheduling optimization results.
[0011] In one possible implementation of this application, the initial weights of different dimensions are adjusted based on the scheduling execution result and production cost variables to obtain adjusted weights, including:
[0012] Based on the scheduling execution results and real-time workshop production parameters, determine the first weight adjustment amount for each dimension;
[0013] Based on the production cost variable and the initial weights, determine the second weight adjustment for each dimension;
[0014] The initial weights of different dimensions are adjusted using the first and second weight adjustment amounts to obtain the adjusted weights.
[0015] In one possible implementation of this application, based on the scheduling execution results and real-time shop floor production parameters, the first weight adjustment amount for each dimension is determined, including:
[0016] Determine the target values for workshop production parameters in each dimension;
[0017] Based on the target value and real-time workshop production parameters, the parameter performance values are calculated.
[0018] Based on the parameter performance values and initial weights, determine the first weight adjustment amount for each dimension.
[0019] In one possible implementation of this application, the second weight adjustment amount for each dimension is determined based on the production cost variable and the initial weights, including:
[0020] Obtain the range of production cost fluctuations;
[0021] Based on the production cost variables and the range of production cost fluctuations within the current preset period, the intensity of production cost changes is calculated.
[0022] The second weight adjustment amount for each dimension is determined by multiplying the intensity of changes in production costs by the initial weights.
[0023] In one possible implementation of this application, the initial weights of different dimensions are adjusted using a first weight adjustment amount and a second weight adjustment amount to obtain adjusted weights, including:
[0024] Calculate the first sum between the first weight adjustment and the second weight adjustment;
[0025] The adjusted weights are calculated based on the first sum and the initial weights of different dimensions.
[0026] In one possible implementation of this application, a correction coefficient is determined based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period, including:
[0027] Obtain target data sequences of different dimensions and dimensional data weight sequences within multiple weight adjustment periods. The target data sequences are used to characterize the target values achieved by the workshop production parameters.
[0028] Calculate the first covariance between the target data sequence and the dimension data weight sequence across different dimensions, and the second covariance between the target data sequence and the dimension data weight sequence within each dimension;
[0029] The degree of conflict in adjusting the weights of different dimensions is determined based on the ratio between the first covariance and the second covariance.
[0030] Based on the adjustment of the degree of conflict, the correction coefficients for each dimension are calculated.
[0031] In one possible implementation of this application, correction coefficients for each dimension are calculated based on the degree of conflict adjustment, including:
[0032] Determine the number of dimensions corresponding to each dimension;
[0033] Based on the number of dimensions and the degree of adjustment conflict between dimensions, the correction coefficients for each dimension are calculated.
[0034] In one possible implementation of this application, the adjusted weights are corrected based on a correction coefficient to obtain corrected weights, including:
[0035] Multiply the sum of the first and second weight adjustments in the adjusted weights by the correction coefficient to obtain the weight correction value;
[0036] The corrected weights are obtained by summing the weight correction value with the initial weights.
[0037] In one possible implementation of this application, the workshop production parameters are scheduled and optimized using modified weights to obtain the scheduling optimization result, including:
[0038] The corrected weights for different dimensions are matched with production data for different dimensions in the workshop production parameters;
[0039] After matching is complete, the corrected weights of each dimension are multiplied by the corresponding production data to obtain the scheduling optimization result.
[0040] This application also provides a multi-objective collaborative scheduling optimization system for intelligent manufacturing workshops. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops as described above.
[0041] This application provides a multi-objective collaborative scheduling optimization method and system for intelligent manufacturing workshops. Compared to related technologies that use a fixed-weight linear summation method, which weights and sums various objective indicators with corresponding weights, the method cannot adjust weights in a timely manner when facing order fluctuations or sudden changes in production energy prices, resulting in poor workshop scheduling optimization. In this application, the method obtains the scheduling execution results of workshop production parameters according to initial weights within a preset time period. Based on the scheduling execution results and production cost variables, the initial weights of different dimensions are adjusted to obtain the adjusted weights. Based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period, a correction coefficient is determined, and then... The adjusted weights are corrected using a correction coefficient to obtain corrected weights. Then, the workshop production parameters are scheduled and optimized using the corrected weights to obtain the scheduling optimization results. By adjusting the initial weights based on the scheduling execution results corresponding to the initial weights and the dynamic changes in production cost variables, the initial weights of different dimensions are adjusted to obtain the adjusted weights. New scheduling weights are generated dynamically and in a timely manner to adapt to the dynamic environment of industrial production. Then, based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment cycle, a correction coefficient is determined. The adjusted weights are corrected based on the correction coefficient to obtain the corrected weights. In this way, accurate adjustment weights of different dimensions are obtained, improving the multi-objective scheduling optimization effect of the workshop. Attached Figure Description
[0042] Figure 1 This is a flowchart illustrating the first embodiment of the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops according to this application.
[0043] Figure 2 This is a flowchart illustrating the second embodiment of the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops in this application.
[0044] Figure 3 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application. Detailed Implementation
[0045] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.
[0046] This application provides a multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops. In the first embodiment of this application's multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops, refer to... Figure 1 The methods include:
[0047] Step S10: Obtain the scheduling execution result obtained by scheduling the workshop production parameters according to the initial weights within a preset time period;
[0048] Step S20: Based on the scheduling execution results and production cost variables, adjust the initial weights of different dimensions to obtain the adjusted weights;
[0049] Step S30: Determine the correction coefficient based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period;
[0050] Step S40: Correct the adjusted weights based on the correction coefficient to obtain the corrected weights;
[0051] Step S50: Optimize the scheduling of workshop production parameters using the corrected weights to obtain the scheduling optimization results.
[0052] This embodiment aims to: adjust the initial weights of different dimensions based on the scheduling execution results corresponding to the initial weights and the dynamic changes of production cost variables to obtain the adjusted weights; dynamically and timely generate new scheduling weights to adapt to the dynamic environment of industrial production; determine the correction coefficient based on the degree of adjustment conflict between the weights of different dimensions within each weight adjustment cycle; and correct the adjusted weights based on the correction coefficient to obtain the corrected weights. In this way, accurate adjustment weights of different dimensions are obtained, thereby improving the multi-objective scheduling optimization effect in the workshop.
[0053] The specific steps are as follows:
[0054] Step S10: Obtain the scheduling execution result obtained by scheduling the workshop production parameters according to the initial weights within a preset time period.
[0055] As an example, the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops can be applied to the multi-objective collaborative scheduling optimization device for intelligent manufacturing workshops, which belongs to the multi-objective collaborative scheduling optimization equipment for intelligent manufacturing workshops.
[0056] As an example, the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops can also be applied to the multi-objective collaborative scheduling optimization system for intelligent manufacturing workshops. In this system, the production parameters of the workshop are optimized by constructing a multi-objective collaborative objective function. This part is prior art, and this application only briefly describes it and does not elaborate on it as a core point. This application divides the objectives of multi-objective optimization into three categories, namely energy consumption, timeliness, and quality. Implementers can also adjust them according to their own circumstances.
[0057] The objective functions for different objectives are:
[0058]
[0059] Parameter description:
[0060] M represents the number of machines, and N represents the number of workpieces;
[0061] x ik ∈{0,1}: Workpiece i is processed on machine k (1 yes / 0 no);
[0062] s i : Start time of workpiece i;
[0063] v k Machine speed (percentage %)
[0064] d i Delivery date of workpiece i;
[0065] p ik : Standard processing time for workpiece i on machine k;
[0066] q ik : The expected defect rate of workpiece i on machine k;
[0067] α k The energy consumption coefficient of machine k;
[0068] β k Machine standby power consumption (k);
[0069] γ, η: mass-velocity coupling coefficients;
[0070] T: Scheduling cycle duration;
[0071] Multi-objective optimization function:
[0072] min(f(x,s,v))
[0073] f(x,s,v)=K1·f1+K2·f2+K3·f3
[0074] Where f1, f2, and f3 represent the objective functions of energy consumption, timeliness, and quality, respectively; K1, K2, and K3 represent the weights of energy consumption, timeliness, and quality, respectively. In related technologies, the initial values of these weights are given through expert consultation or based on empirical values. This application addresses the fixed weight problem in existing workshop multi-objective collaborative scheduling optimization by proposing a closed-loop feedback adjustment scheme for multi-objective weights. By combining the scheduling execution results within a certain period and the dynamic changes of environmental factors, the scheduling weights of different dimensions of data are corrected, thereby optimizing the scheduling based on the corrected weights to obtain better scheduling optimization results.
[0075] As an example, the preset time period can be 5 days, 10 days, etc. in the historical production process, and there is no specific limitation. The workshop production parameters are energy consumption, timeliness and quality, which represent several types of parameters of different dimensions generated in the workshop production process. When performing scheduling optimization, these target parameters need to be weighed with different weights.
[0076] As an example, the scheduling execution result can be the result obtained after scheduling based on a fixed initial weight in the relevant technology. After collecting historical scheduling execution results, the initial weight is adjusted by combining historical data and considering the degree of change of various production costs in the actual environment.
[0077] Step S20: Based on the scheduling execution results and production cost variables, adjust the initial weights of different dimensions to obtain the adjusted weights.
[0078] As an example, production cost variables are the actual cost variables consumed. Taking energy consumption as an example, production cost variables can be energy consumption variables.
[0079] As an example, the scheduling execution results within a historical time period are combined with the actual production cost variables to dynamically adjust the initial weights of different dimensions, resulting in the adjusted weights.
[0080] As an example, the adjusted weights are obtained by calculating two weight adjustment amounts. Taking energy consumption as an example, the first weight adjustment amount can be calculated based on the difference between actual energy consumption and target energy consumption, and the second weight adjustment amount can be calculated based on the degree of environmental change for different target values. The initial weights are adjusted based on these two weight adjustment amounts.
[0081] Among them, step S20 of the multi-objective collaborative scheduling optimization in the intelligent manufacturing workshop also includes steps S21 to S23, including:
[0082] Step S21: Based on the scheduling execution results and real-time workshop production parameters, determine the first weight adjustment amount for each dimension.
[0083] As an example, real-time workshop production parameters include real-time energy consumption, real-time efficiency, and real-time quality. Based on the target value corresponding to the scheduling execution result and the real-time workshop production parameters, the difference between the actual parameters and the target parameters is determined, thereby determining the weight adjustment amount that needs to be calculated, that is, the first weight adjustment amount.
[0084] Step S21, which determines the first weight adjustment amount for each dimension based on the scheduling execution results and real-time workshop production parameters, includes:
[0085] Determine the target values for workshop production parameters in each dimension.
[0086] As an example, the target value of workshop production parameters can be extracted from the scheduling execution results within a historical time period. The target value can be a fixed value or a range of target data.
[0087] As an example, for an energy consumption target value, production energy consumption data of similar products / processes over the past 30 days can be selected and weighted moving averages can be applied (new data has higher weights, and the weighting method is not limited) to obtain the energy consumption target value E for a certain period. target .
[0088] As an example, for the delivery timeliness target value, the required date in the user contract is obtained through the production management system and used as the delivery timeliness target value D. target .
[0089] As an example, for the quality target value, the defect rate / rejection rate threshold in the quality inspection process can be used as the quality target value Z, either through historical data or by manually setting it. target For example, the defect rate threshold can be set to 20%, with no specific limit.
[0090] Based on the target value and real-time workshop production parameters, the parameter performance values are calculated.
[0091] Based on the parameter performance values and initial weights, determine the first weight adjustment amount for each dimension.
[0092] As an example, the parameter performance values include energy consumption performance values, timeliness performance values, and quality performance values. These parameter performance values are the ratios between real-time workshop production parameters and target values. For instance, the actual energy consumption value E can be calculated using the objective function f1. c Then calculate the energy performance value TE:
[0093]
[0094] Similarly, the timeliness performance value TD and quality performance value TZ of other dimensions are calculated.
[0095] As an example, after calculating the corresponding timeliness performance value, based on the timeliness performance value and the initial weight value, a first weight adjustment amount is calculated based on the difference between the actual parameters and the target parameters. The first weight adjustment amount includes a first energy consumption weight adjustment amount ΔK. E First time-sensitive weight adjustment amount ΔKT D First mass weight adjustment amount ΔKT Z .
[0096] Taking energy consumption data as an example, the first energy consumption weight adjustment amount ΔKT E The calculation method can be:
[0097] ΔKT E =K E ×(1-TE)
[0098] Where: ΔKT EThis represents the adjustment amount of the energy consumption weight based on data performance, that is, the adjustment amount of the first energy consumption weight; TE represents the energy consumption performance value; K E The initial weights for energy consumption can be represented, and similarly, the first time-dependent weight adjustment ΔKT can be calculated. D and the first mass weight adjustment amount ΔKT Z .
[0099] If the data in the current dimension performs better, it means that the data in the current dimension is more in line with expectations. At this time, the weight can meet the target requirements, and the weight should be smaller. Conversely, the weight should be larger. The weight adjustment amount can be negative. When the adjustment amount is negative, the weight of the corresponding data decreases.
[0100] Step S22: Based on the production cost variable and the initial weights, determine the second weight adjustment amount for each dimension.
[0101] As an example, production cost variables can also be the electricity consumption generated during the production process. For instance, when electricity prices rise, the costs consumed during the production process will increase under the same conditions. The initial weight is the weight value set at the beginning, usually based on empirical values.
[0102] As an example, the second weight adjustment amount can be a weight adjustment amount calculated based on the degree of change in production costs for different target values, and the weight can be dynamically adjusted in combination with real-time cost changes.
[0103] Step S22, which determines the second weight adjustment for each dimension based on the production cost variable and the initial weights, includes:
[0104] Obtain the range of production cost fluctuations.
[0105] Based on the production cost variables and the range of production cost fluctuations within the current preset period, the intensity of production cost changes is calculated.
[0106] As an example, the definition of the production cost fluctuation range differs for different dimensions of data. For instance, for energy consumption, the production cost fluctuation range could be the electricity price fluctuation range; for timeliness, the production cost fluctuation range could be the time limit buffer fluctuation range; and for quality, the production cost fluctuation range could be the equipment health index range or the health index threshold. In this embodiment, the health index threshold is used as an example. When the equipment health index exceeds the health index threshold, it indicates that the equipment has a significant risk and needs to be shut down.
[0107] As an example, the preset period can be 1 minute or 2 minutes. Based on the production cost variables and the range of production cost fluctuations within the current preset period, the intensity of production cost change is calculated.
[0108] As an example, the intensity of changes in production costs can be the intensity of changes in energy consumption, the intensity of changes in timeliness, and the intensity of changes in quality, specifically:
[0109] The energy consumption change intensity pv can be calculated as follows:
[0110]
[0111] In the formula: p represents the current electricity price, which is obtained through the power grid API interface; p0 represents the electricity price before the change; p r This indicates the range of electricity price fluctuations, and it is calculated by subtracting the average electricity price from the peak price.
[0112] The calculation method for the intensity of change over time (dv) can be:
[0113]
[0114] In the formula: d represents the total construction period consumed under the current time limit, and the updated order time limit d is obtained from the system database; D target This represents the target timeframe, i.e., the estimated construction period; d r This indicates the time limit buffer range, which is obtained by the average difference between the actual delivery date and the target delivery date of historical orders, or it can be set manually.
[0115] The intensity of mass change zv can be calculated as follows:
[0116]
[0117] In the formula: z represents the current health index of the device, which is obtained by combining IoT sensors (vibration meter / thermocouple) with edge computing to obtain the change in the device health index (the health index is calculated using existing device health index methods; any existing technology can be used, for example, device health index = 0.7 × vibration value + 0.3 × temperature); z0 represents the health index of the device in the previous planning cycle; z m This represents the health index threshold. Exceeding this health index indicates a significant risk to the equipment, requiring shutdown. The risk index of production equipment indirectly reflects the quality of the workpieces it produces, and thus, the corresponding quality target weight increment can be calculated.
[0118] The second weight adjustment amount for each dimension is determined by multiplying the intensity of changes in production costs by the initial weights.
[0119] As an example, the second weight adjustment includes the second energy consumption weight adjustment, the second timeliness weight adjustment, and the second quality weight adjustment.
[0120] Taking energy consumption as an example, the second energy consumption weight adjustment amount ΔKV EThe calculation method can be:
[0121] ΔKV E =K E ×pv
[0122] Where: ΔKV E K represents the second energy consumption weighting adjustment based on the production cost variable; E The initial weight of energy consumption is represented by pv; the intensity of energy consumption change is represented by pv. Similarly, the second time-dependent weight adjustment ΔKV can also be calculated. D Second mass weight adjustment amount ΔKV Z .
[0123] Step S23: Adjust the initial weights of different dimensions using the first weight adjustment amount and the second weight adjustment amount to obtain the adjusted weights.
[0124] The step S23, which adjusts the initial weights of different dimensions using the first and second weight adjustment amounts to obtain the adjusted weights, includes:
[0125] Calculate the first sum between the first weight adjustment and the second weight adjustment.
[0126] As an example, the first sum could be the sum of the first weight adjustment and the second weight adjustment.
[0127] The adjusted weights are calculated based on the first sum and the initial weights of different dimensions.
[0128] As an example, the first sum is added to the initial weights of different dimensions to obtain the adjusted weights. In order to ensure that the sum of the weights of the data in different dimensions is 1, each weight needs to be normalized according to the sum of its adjusted weights.
[0129] As an example, the adjusted weight can also be calculated by first adjusting the weight using the first weight adjustment amount, and then adjusting the weight again using the second weight adjustment amount to obtain the adjusted weight. In the second adjustment process, the second weight adjustment amount is calculated based on the weight adjusted in the first adjustment (not the initial weight). The adjustment effects of multiple adjustment cycles can be compared to select the optimal weight adjustment method. No specific limitation is made.
[0130] Step S30: Determine the correction coefficient based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period.
[0131] As an example, when adjusting the weights of different dimensions, the dimensions will have an interrelationship rather than be independent of each other. For example, reducing energy consumption to reduce project delivery delays, the degree of adjustment conflict indicates the extent of the conflict in the weight adjustment between different dimensions. The weight adjustment of each dimension is corrected / suppressed according to the degree of adjustment conflict.
[0132] As an example, the correction coefficient is mainly used to correct the calculated weight adjustment amount, and then the corrected weight is calculated based on the corrected weight adjustment amount and the initial weight.
[0133] Step S40: Correct the adjusted weights based on the correction coefficient to obtain the corrected weights.
[0134] As an example, after calculating the correction coefficient, the adjusted weights are corrected using the correction coefficient to obtain the corrected weights.
[0135] The step S40, which corrects the adjusted weights based on the correction coefficient to obtain the corrected weights, includes:
[0136] The weight correction value is obtained by multiplying the sum of the first and second weight adjustments in the adjusted weights by the correction coefficient.
[0137] As an example, taking the weight adjustment amount corresponding to energy consumption data as an example, the weight correction value is expressed as:
[0138] γ E (ΔKT E +ΔKV E )
[0139] Where, γ E Denotes the correction factor, ΔKT E This represents the first weighted adjustment amount corresponding to energy consumption, ΔKV. E This represents the second weight adjustment amount corresponding to energy consumption; the weight adjustment values for other dimensions are similar.
[0140] The corrected weights are obtained by summing the weight correction value with the initial weights.
[0141] As an example, based on the performance of target data and changes in production cost data, and considering the conflict between different dimensions of data, the weights of each dimension are adjusted. Taking energy consumption data as an example, the corrected weight K′ is... E The calculation method can be:
[0142] K′ E =K E +γ E (ΔKT E +ΔKV E )
[0143] In the formula: K′ E K represents the corrected weight for energy consumption. E The initial weight representing energy consumption; ΔKT E This represents the adjustment amount of energy consumption weight based on data performance; that is, the first weight adjustment amount corresponding to energy consumption; ΔKV E This represents the weighted adjustment based on environmental variables, specifically the second weighted adjustment corresponding to energy consumption; γ E This represents the correction factor for the energy consumption dimension. Similarly, the corrected weights K′ for the timeliness of other dimensions can also be calculated. D The quality-corrected weight K′ Z .
[0144] As an example, the adjustment weights of different dimensions are normalized using the sum of weights to ensure that the sum of weights is 1. Specifically, the weights of the three dimensions are added together, and then the three weights are divided by the sum of weights to obtain the final weight values.
[0145] Step S50: Optimize the scheduling of workshop production parameters using the corrected weights to obtain the scheduling optimization results.
[0146] The scheduling optimization results are obtained by optimizing the workshop production parameters using the corrected weights, including:
[0147] The corrected weights for different dimensions are matched with production data for different dimensions in the workshop production parameters.
[0148] After matching is complete, the corrected weights of each dimension are multiplied by the corresponding production data to obtain the scheduling optimization result.
[0149] As an example, the corrected weights include the corrected weight K′ for energy consumption. E The weight K′ after timeliness correction D The quality-corrected weight K′ Z After calculating these weights, multiply each weight by energy consumption, timeliness, and quality, and then add the products of each weight to obtain the scheduling optimization result.
[0150] As an example, by using the corrected weights and combining them with the objective functions of different dimensions mentioned above, a heuristic optimization algorithm is used to optimize the overall scheme, and then the multi-objective collaborative scheduling of the workshop is carried out based on the optimization results.
[0151] This application provides a multi-objective collaborative scheduling optimization method and system for intelligent manufacturing workshops. Compared to related technologies that use a fixed-weight linear summation method, which weights and sums various objective indicators with corresponding weights, the method cannot adjust weights in a timely manner when facing order fluctuations or sudden changes in production energy prices, resulting in poor workshop scheduling optimization. In this application, the method obtains the scheduling execution results of workshop production parameters according to initial weights within a preset time period. Based on the scheduling execution results and production cost variables, the initial weights of different dimensions are adjusted to obtain the adjusted weights. Based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period, a correction coefficient is determined, and then... The adjusted weights are corrected using a correction coefficient to obtain corrected weights. Then, the workshop production parameters are scheduled and optimized using the corrected weights to obtain the scheduling optimization results. By adjusting the initial weights based on the scheduling execution results corresponding to the initial weights and the dynamic changes in production cost variables, the initial weights of different dimensions are adjusted to obtain the adjusted weights. New scheduling weights are generated dynamically and in a timely manner to adapt to the dynamic environment of industrial production. Then, based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment cycle, a correction coefficient is determined. The adjusted weights are corrected based on the correction coefficient to obtain the corrected weights. In this way, accurate adjustment weights of different dimensions are obtained, improving the multi-objective scheduling optimization effect of the workshop.
[0152] Furthermore, referring to Figure 2 Based on the first embodiment of this application, another embodiment of this application is provided. In this embodiment, step S30, which determines the correction coefficient based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period, includes steps S31 to S34:
[0153] Step S31: Obtain target data sequences of different dimensions and dimension data weight sequences within multiple weight adjustment cycles. The target data sequences are used to characterize the target values achieved by the workshop production parameters.
[0154] As an example, after calculating the first and second weight adjustment amounts, the initial weights are adjusted using these two weight adjustment amounts. The time period during the weight adjustment process is the weight adjustment period, which can be 1 hour or 5 hours, without any specific limitation.
[0155] As an example, the target data sequence is a sequence of target values of workshop production parameters obtained after adjusting the weights within different weight adjustment cycles. For each dimension, there is a corresponding target data sequence, such as the energy consumption data sequence.
[0156] As an example, a dimensional data weight sequence can be a sequence of weights for the same dimension with different weight adjustment periods.
[0157] Step S32: Calculate the first covariance between the target data sequence and the dimension data weight sequence in different dimensions, and the second covariance between the target data sequence and the dimension data weight sequence in each dimension.
[0158] As an example, the first covariance can be the covariance between the target data sequence and the dimension data weight sequence across different dimensions. Taking the i-th dimension and the k-th dimension as an example, the first covariance S(i,k) can be calculated as follows:
[0159] S(i,k)=cov(F i,j ,K k,j )
[0160] Where i and k represent different dimension indices, F i,j Let K represent the target data sequence in the i-th dimension and the j-th weight adjustment period. k,j This represents the dimensional data weight sequence for the k-th dimension and the j-th weight adjustment period.
[0161] As an example, taking the i-th dimension as an example, the second covariance S(i,i) can be calculated as follows:
[0162] S(i,i)=cov(F i,j ,K i,j )
[0163] Among them, F i,j Let K represent the target data sequence in the i-th dimension and the j-th weight adjustment period. i,j This represents the dimensional data weight sequence for the i-th dimension and the j-th weight adjustment period.
[0164] As an example, the first covariance and the second covariance are used to represent the positive or negative correlation between two sequences. When the covariance is positive, it is a positive correlation, which is reflected in the two sequences increasing or decreasing at the same time. When the covariance is negative, it is a negative correlation, with one sequence increasing and the other decreasing.
[0165] Step S33: Based on the ratio between the first covariance and the second covariance, determine the degree of conflict in the adjustment of weights between different dimensions.
[0166] As an example, since scheduling optimization methods are based on weights to weigh different dimensions, for a single dimension, its target data and its weight should generally show a strong negative correlation. The stronger the conflict between two dimensions, the more the target data of the other dimension will increase when the weight of one dimension increases. That is, the stronger the correlation between the data sequence and the weight of the two dimensions, the stronger the conflict.
[0167] As an example, the degree of conflict in adjusting weights between different dimensions, Q(i,k), can be calculated as follows:
[0168]
[0169] In the formula: Q(i,k) represents the degree of adjustment conflict between dimension k and dimension i; S(i,k) represents the first covariance between the target data sequence of dimension i and dimension k; S(i,i) represents the second covariance between the target data sequence of dimension i and its own weight data.
[0170] Step S34: Based on the degree of conflict adjustment, calculate the correction coefficients for each dimension.
[0171] As an example, based on the degree of conflict between different dimensions, the correction coefficients of each dimension are calculated, thereby correcting the weights of different target data to a reasonable range or adjusting them to a reasonable weight value.
[0172] Step S34, which calculates the correction coefficients for each dimension based on the adjusted conflict level, includes:
[0173] Determine the number of dimensions corresponding to each dimension.
[0174] Based on the number of dimensions and the degree of adjustment conflict between dimensions, the correction coefficients for each dimension are calculated.
[0175] As an example, the number of dimensions refers to the number of dimensions corresponding to energy consumption, timeliness, and quality. In this embodiment, the number of dimensions is 3.
[0176] As an example, the inhibition coefficient of each dimension is calculated based on its conflict with other dimensions:
[0177]
[0178] Wherein: γ k Q(i,k) represents the correction coefficient for dimension k; Q(i,k) represents the degree of adjustment conflict between dimension k and dimension i; i represents the index of each dimension, and n is the number of dimensions, which is 3 in this application.
[0179] In this embodiment, the correction coefficients for each dimension are calculated based on the degree of conflict between each dimension and other dimensions, and the weight adjustment amount is corrected using the correction coefficients to determine the accurate weight value.
[0180] Specifically, this application also provides a multi-objective collaborative scheduling optimization system for intelligent manufacturing workshops. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops as described above.
[0181] Reference Figure 3 , Figure 3 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of this application.
[0182] like Figure 3 As shown, the multi-objective collaborative scheduling optimization device in this intelligent manufacturing workshop may include: a processor 1001, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the processor 1001 and the memory 1005.
[0183] Optionally, the multi-objective collaborative scheduling and optimization equipment in this intelligent manufacturing workshop may also include a user interface, a network interface, a camera, RF (Radio Frequency) circuits, sensors, a WiFi module, etc. The user interface may include a display screen and an input submodule such as a keyboard; optional user interfaces may also include standard wired or wireless interfaces. The network interface may include standard wired or wireless interfaces (such as a Wi-Fi interface).
[0184] Those skilled in the art will understand that Figure 3 The multi-objective collaborative scheduling optimization equipment structure shown in the figure does not constitute a limitation on the multi-objective collaborative scheduling optimization equipment in the intelligent manufacturing workshop. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0185] like Figure 3 As shown, the memory 1005, serving as a storage medium, may include an operating system, a network communication module, and a multi-objective collaborative scheduling optimization program for the intelligent manufacturing workshop. The operating system is a program that manages and controls the hardware and software resources of the multi-objective collaborative scheduling optimization equipment in the intelligent manufacturing workshop, supporting the operation of the multi-objective collaborative scheduling optimization program and other software and / or programs. The network communication module is used to enable communication between the various components within the memory 1005, as well as communication with other hardware and software in the multi-objective collaborative scheduling optimization system of the intelligent manufacturing workshop.
[0186] exist Figure 3In the multi-objective collaborative scheduling optimization device for the intelligent manufacturing workshop shown, the processor 1001 is used to execute the multi-objective collaborative scheduling optimization program for the intelligent manufacturing workshop stored in the memory 1005, and implement the steps of the multi-objective collaborative scheduling optimization method for the intelligent manufacturing workshop mentioned above.
[0187] The specific implementation method of the multi-objective collaborative scheduling optimization equipment for intelligent manufacturing workshops in this application is basically the same as the embodiments of the multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops described above, and will not be repeated here.
[0188] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.
[0189] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0190] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of this application.
[0191] The above are merely preferred embodiments of this application and do not limit the scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.
[0192] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0193] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.
Claims
1. A multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops, characterized in that, The method includes: Obtain the scheduling execution results obtained by scheduling workshop production parameters according to initial weights within a preset time period; Based on the scheduling execution results and production cost variables, the initial weights of different dimensions are adjusted to obtain the adjusted weights, specifically including: Determine the target values for workshop production parameters in each dimension, wherein the target values are extracted from the scheduling execution results; Based on the target value and real-time workshop production parameters, parameter performance values are calculated, including energy consumption performance values, timeliness performance values, and quality performance values. Based on the parameter performance values and the initial weights, determine the first weight adjustment amount for each dimension; Obtain the range of production cost fluctuations; Based on the production cost variables within the current preset period and the production cost fluctuation range, the intensity of production cost change is calculated, which includes the intensity of energy consumption change, the intensity of timeliness change, and the intensity of quality change. The second weight adjustment amount for each dimension is determined based on the product of the intensity of the change in production cost and the initial weight. The initial weights of different dimensions are adjusted using the first weight adjustment amount and the second weight adjustment amount to obtain the adjusted weights; Based on the degree of adjustment conflict between weights of different dimensions within each weight adjustment period, a correction coefficient is determined, specifically including: Obtain target data sequences of different dimensions and dimensional data weight sequences within multiple weight adjustment cycles. The target data sequences are used to characterize the target values achieved by workshop production parameters. Calculate the first covariance between the target data sequence and the dimension data weight sequence across different dimensions, and the second covariance between the target data sequence and the dimension data weight sequence within each dimension; The degree of conflict in adjusting the weights of different dimensions is determined based on the ratio between the first covariance and the second covariance. Determine the number of dimensions corresponding to the given dimension; Based on the number of dimensions and the degree of adjustment conflict between each dimension, the correction coefficient for each dimension is calculated. The adjusted weights are corrected based on the correction coefficient to obtain the corrected weights; The workshop production parameters are scheduled and optimized using the corrected weights to obtain the scheduling optimization results.
2. The multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops as described in claim 1, characterized in that, The step of adjusting the initial weights of different dimensions using the first weight adjustment amount and the second weight adjustment amount to obtain the adjusted weights includes: Calculate the first sum between the first weight adjustment and the second weight adjustment; Based on the first sum and the initial weights of different dimensions, the adjusted weights are calculated.
3. The multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops as described in claim 1, characterized in that, The step of correcting the adjusted weights based on the correction coefficient to obtain the corrected weights includes: The sum of the first weight adjustment and the second weight adjustment in the adjusted weights is multiplied by the correction coefficient to obtain the weight correction value; The corrected weights are obtained by summing the corrected weights with the initial weights.
4. The multi-objective collaborative scheduling optimization method for intelligent manufacturing workshops as described in claim 1, characterized in that, The step of optimizing the workshop production parameters using the corrected weights to obtain the optimization result includes: The corrected weights for different dimensions are matched with production data for different dimensions in the workshop production parameters; After matching is complete, the corrected weights of each dimension are multiplied by the corresponding production data to obtain the scheduling optimization result.
5. A multi-objective collaborative scheduling and optimization system for an intelligent manufacturing workshop, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that the processor executes the computer program to implement the steps of the method as described in any one of claims 1 to 4.