Energy Storage Cabinet Manufacturing, Production, Testing, Energy Consumption Collaborative Scheduling Management System and Method

By coupling the scheduling model of production cycle, testing station and energy supply, and combining genetic algorithm and phase flexible adjustment, the problem of high electricity purchase cost and peak demand caused by independent scheduling of energy storage equipment production and testing is solved, and the efficiency of energy mutual assistance between equipment is improved and resources are optimized.

CN123117734APending Publication Date: 2026-11-17FUZHOU UNIV
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202611618219.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The independent scheduling of production and charging/discharging testing for existing energy storage equipment leads to high electricity purchase costs, large peak demand, and untapped charging/discharging potential between equipment.

Method used

By constructing a scheduling model that couples production cycle time, testing station and energy supply, and solving it using a genetic algorithm, combined with flexible adjustment of test path phase, the configuration of production line and testing station is optimized to achieve linkage between production plan and test plan, and improve the mutual assistance efficiency of charging and discharging energy.

Benefits of technology

It significantly reduces electricity purchase costs and peak demand across the plant, optimizes resource allocation for production lines and testing stations, increases the overlap of charging and discharging times between equipment, and reduces grid waste electricity. It is suitable for small-batch and large-scale energy storage cabinet production and testing scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117734A_ABST
    Figure CN123117734A_ABST
Patent Text Reader

Abstract

This invention discloses a collaborative scheduling management system and method for energy consumption in energy storage cabinet manufacturing, production, and testing, belonging to the field of energy storage equipment production scheduling and energy management technology. Addressing the problems of high electricity purchase costs, large peak demand, fixed test path timing, and insufficient utilization of charging / discharging support potential between equipment caused by independent scheduling of production and charging / discharging testing, this invention constructs a scheduling model coupling production cycle time, testing stations, and energy support. Using the number of production lines, the number of testing stations, equipment testing priority, and testing paths as decision variables, and aiming to minimize the sum of grid electricity purchase costs and equipment operating costs, a genetic algorithm is employed to solve the problem. Furthermore, a flexible phase adjustment mechanism for the testing path is introduced, optimizing the charging / discharging window matching through inter-stage buffer time. This invention can improve energy support rate, reduce electricity purchase costs and peak demand, balance equipment investment, and adapt to different scale production scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the cross-technical field of energy storage equipment production scheduling and energy management, specifically to a collaborative scheduling management system and method for energy consumption in energy storage cabinet manufacturing, production, and testing. Background Technology

[0002] Energy costs in the testing phase have become a significant component of the manufacturing cost of energy storage equipment. Currently, industrial users generally implement time-of-use pricing and peak demand pricing policies, with a large price difference between peak and off-peak periods. Demand charges are based on the maximum demand within the billing cycle, and the timing arrangement of the test load directly determines the overall electricity cost of the factory.

[0003] In existing technologies, the production scheduling and charging / discharging testing of energy storage devices are mostly independently managed, belonging to two separate business management processes, and have not yet achieved coordinated optimization of production resources, testing resources, and on-site energy. Existing related technologies can be mainly divided into two major directions: production scheduling optimization and testing energy optimization.

[0004] Category 1: Production scheduling technologies for batteries and energy storage products This type of technology focuses on production line capacity constraints and order delivery cycles, with the core objectives of improving production efficiency and ensuring delivery cycles.

[0005] Chinese patent application CN119784099A discloses a collaborative scheduling method for multiple battery production lines. It constructs a multi-objective production optimization model aimed at minimizing product delay time and process changeover time. Using an alternating iterative optimization mathematical model and heuristic algorithms, it adjusts processes to shorten product completion time and reduce order delivery delays. The advantage of this solution is its ability to optimize resource allocation across multiple parallel production lines, ensuring order delivery cycles and supporting dynamic scheduling adjustments in abnormal scenarios such as material inventory warnings and equipment failures. However, this patent only schedules production processes and does not consider the power load and electricity costs associated with subsequent charging and discharging testing. Production and testing plans are disconnected, failing to achieve joint optimization of schedule and energy costs from a plant-wide perspective.

[0006] Chinese patent application CN120087729A discloses a deep learning-based scheduling method for MES systems in the lithium battery industry. This method integrates a dual-tower Siamese network, Transformer-XL long-sequence modeling, and a neural Turing machine to construct an intelligent scheduling model. It can uncover long-cycle production dependencies, improving the intelligence level of production scheduling and its dynamic adaptability to production disturbances. The advantage of this technology is its ability to handle dynamic disturbances on the production line and output real-time scheduling solutions. However, its optimization boundary is also limited to the manufacturing process, not relating to the back-end charging and discharging testing processes, and cannot account for the energy consumption during the testing phase.

[0007] Category 2: Energy optimization technologies for charge and discharge testing of energy storage products Since conventional enterprises do not feed back into the grid, the discharge energy generated during the testing of energy storage equipment cannot be converted into electricity revenue if it is directly fed back to the grid. Instead, it will increase the risk of grid-side management and energy loss. Therefore, this type of technology focuses on energy recovery and mutual assistance within the testing station, mainly to solve the problems of energy waste and grid impact during the testing process.

[0008] Chinese patent application CN119199621A discloses a test method and test platform for towing tests of prefabricated battery modules. It constructs a charge-discharge test loop between two prefabricated modules through a converter control system, achieving internal energy circulation for testing, reducing the impact on the power grid during the test process, and minimizing energy waste during discharge. This scheme can achieve energy towing under limited workstation pairing conditions, reducing the amount of electricity fed into the grid. However, this scheme only supports static pairing tests at fixed workstations, does not consider the test time constraints caused by upstream production cycle time, and cannot dynamically adjust the test sequence according to production completion time. Furthermore, the model does not incorporate industrial electricity price cost factors such as time-of-use pricing and peak demand charges, making it unable to achieve full-cycle cost minimization scheduling for large-scale, multi-model equipment.

[0009] Chinese patent CN116365671B discloses a PCS testing system and method for energy storage containers. By calculating excess electricity during the testing process, it limits the power fed into the grid, reducing the impact on the power grid and minimizing energy waste during discharge. This solution can suppress the problem of high-power energy feeding back into the grid under testing conditions. However, this solution only manages power for a single set or small batch of testing equipment, lacking global scheduling logic for multiple workstations and devices across the entire plant. It does not consider production completion time as a constraint, nor does it incorporate time-of-use pricing and demand charges into the optimization objectives.

[0010] In summary, current technologies generally suffer from the drawback of fragmented scheduling between the manufacturing and charging / discharging testing stages. On the one hand, this results in random superposition of testing loads over time, leading to persistently high peak power consumption across the plant and consequently, high peak-hour and demand-based electricity costs. On the other hand, the inherent energy mutual support potential of parallel testing of multiple devices is not fully exploited, with equipment discharge energy frequently fed back to the grid, resulting in energy waste. Therefore, there is an urgent need in this field for a multi-stage collaborative scheduling scheme that simultaneously couples production cycle constraints, testing station constraints, and energy mutual support constraints. This scheme should fully leverage the energy mutual support capabilities between devices while ensuring order delivery requirements, thereby simultaneously reducing plant electricity costs and peak demand. Summary of the Invention

[0011] The technical problem this invention aims to solve is that the existing independent scheduling of production and charging / discharging testing leads to high electricity purchase costs, large peak demand, and unutilized charging / discharging capacity between equipment. To address these issues, this invention provides a collaborative scheduling management system and method for energy consumption in energy storage cabinet manufacturing and testing. Through coupled scheduling of production cycle time, testing stations, and energy consumption, along with flexible adjustment of testing path phases, it significantly improves the efficiency of energy exchange during charging and discharging while ensuring delivery schedules, reducing electricity purchase costs and overall plant peak demand. Simultaneously, it optimizes the configuration of production lines and testing stations to balance equipment investment and adapts to energy storage cabinet production and testing scenarios of different scales.

[0012] To achieve the above objectives, the present invention adopts the following technical solution.

[0013] A collaborative scheduling and management method for energy consumption in the manufacturing, production, and testing of energy storage cabinets includes the following steps: S1: Obtain order information and equipment parameters, including equipment model, quantity of each model, production time of a single unit, energy storage capacity, initial state of charge, and selectable test charging and discharging paths; S2: Construct a scheduling model coupling production cycle time, testing stations, and energy support. This model includes production cycle time constraints, testing station constraints, energy support constraints, and an objective function. The number of production lines and testing stations are decision variables to be optimized. The production cycle time constraint ensures that each piece of equipment sequentially enters multiple parallel production lines for assembly, and the testing start time of each piece of equipment is no earlier than its production completion time. The testing station constraint ensures that each piece of equipment occupies one testing station for charging and discharging testing, and the same testing station can accommodate at most one piece of equipment at any given time. The energy support constraint ensures that during the testing phase, each piece of equipment generates minute-level charging and discharging power curves according to the selected charging and discharging path. The total net power of the plant is the sum of the power of each piece of equipment, with priority given to offsetting charging power support. The objective function minimizes the sum of grid power purchase cost, equipment operating cost, and completion timeout penalty within the scheduling cycle. The grid power purchase cost includes time-of-use electricity charges and peak demand charges. S3: The scheduling model is solved using a genetic algorithm: the number of production lines, the number of test stations, the test priority of each device, and the test path constitute the encoded individuals. During decoding, production and testing are arranged for each device in order of priority, a scheduling scheme is generated, and the objective function value is calculated as the individual fitness. S4: Output the optimal scheduling scheme, which includes the production start and end times, test start and end times, test station numbers, test paths, and the optimal number of production lines and the optimal number of test stations for each device.

[0014] Furthermore, the time-of-use electricity fee is calculated by multiplying the net power purchased per minute within the scheduling cycle by the corresponding minute-based time-of-use electricity price; the peak demand electricity fee is the product of the peak net power purchased within the scheduling cycle and the demand electricity price; the equipment operating cost is the product of the number of production lines and the operating cost of a single production line, plus the product of the number of test stations and the operating cost of a single test station. The operating cost of a single production line is calculated by dividing the purchase cost of the production line by its service life, and the operating cost of a single test station is calculated by dividing the purchase cost of the test station by its service life; the completion timeout penalty is the portion of the scheduling scheme's completion time exceeding the scheduling cycle multiplied by a unit time penalty coefficient.

[0015] Furthermore, the production process arrangement rules during decoding are as follows: equipment is selected in order of priority and assigned to the earliest available production line. The completion time of equipment production is the sum of the release time of the production line and the equipment production time.

[0016] Furthermore, the arrangement rules for the testing phase during decoding are as follows: the earliest testable time is the time when the equipment is completed. Starting from that time, the available time periods for each test station are searched in a sliding window within the scheduling cycle at fixed steps, and the test start time and test station that minimizes incremental costs are selected.

[0017] Furthermore, the incremental cost includes incremental time-of-use electricity charges, incremental demand charges, peak limit penalties, and waste electricity penalties; wherein, the peak limit penalty is calculated by multiplying the excess amount by a penalty coefficient when the peak value of the global net purchased power exceeds the preset peak limit; the waste electricity penalty is the penalty corresponding to the portion of the discharge power that is not offset by the grid and is fed back to the grid.

[0018] Furthermore, the specific rules for the priority of discharge power and the offsetting of charging power are as follows: within the same minute, the discharge power of each test device is given priority to supply the charging needs of each test device within that minute. The portion of discharge power exceeding the charging needs is recorded as waste power, and the portion of charging needs exceeding discharge power is supplemented by electricity purchased from the power grid.

[0019] Furthermore, the encoded individual consists of a production line quantity gene, a test station quantity gene, N test priority weights in the first segment, and N test path selections in the second segment, where N is the total number of devices; the genetic operations include two-point crossover, gene position probability mutation, and tournament selection, and adopt a historical best individual retention strategy; during decoding, the device processing order is determined according to the priority weight in ascending order; the mutation method of the production line quantity gene and the test station quantity gene is to add or subtract 1 on the original value and limit it to their respective value range, the priority bit mutation is to generate a random real number, and the path bit mutation is to switch to another test path.

[0020] Furthermore, the test charge and discharge path includes the following two types: Path 1: charge from the initial state of charge to 100%, then discharge to 0%, and then recharge to 30%; Path 2: discharge from the initial state of charge to 0%, then charge to 100%, then discharge to 0%, and then recharge to 30%.

[0021] The energy storage cabinet manufacturing, production, testing, and energy consumption collaborative scheduling management system includes: The parameter acquisition module is used to acquire order information and equipment parameters. The equipment parameters include equipment model, quantity of each model, production time of a single unit, energy storage capacity, initial state of charge, and selectable test charging and discharging paths. The model building module is used to construct a scheduling model coupling production cycle time, testing stations, and energy supply constraints. This module includes a production cycle time constraint unit, a testing station constraint unit, an energy supply constraint unit, and an objective function unit. The number of production lines and testing stations are the decision variables to be optimized. The production cycle time constraint unit limits each piece of equipment to sequentially enter multiple parallel production lines for assembly production, and the start time of equipment testing is no earlier than its production completion time. The testing station constraint unit limits each piece of equipment to occupy one testing station for charging and discharging testing, and the same testing station can accommodate at most one piece of equipment at the same time period. The energy supply constraint unit limits each piece of equipment during the testing phase to generate minute-level charging and discharging power curves according to the selected charging and discharging path. The total net power of the plant is the sum of the power of each piece of equipment, with discharging power preferentially offsetting charging power. The objective function unit aims to minimize the sum of grid power purchase cost, equipment operating cost, and completion timeout penalty within the scheduling cycle. The grid power purchase cost includes time-of-use electricity charges and peak demand charges. An optimization solution module is used to solve the scheduling model using a genetic algorithm. This module includes an encoding unit, a decoding unit, a genetic operation unit, and a fitness evaluation unit. The encoding unit is used to construct encoded individuals based on the number of production lines, the number of test stations, the test priority of each device, and the test path. The decoding unit is used to arrange production and testing for each device in order of priority to generate a scheduling scheme. The fitness evaluation unit is used to calculate the objective function value as the individual's fitness. A scheme output module is used to output the optimal scheduling scheme, which includes the production start and end times, test start and end times, test station numbers, test paths, and the optimal number of production lines and test stations for each device.

[0022] Furthermore, the rules for scheduling the test phase by the decoding unit are as follows: taking the equipment production completion time as the earliest testable time, starting from that time, the available time period of each test station is searched in a sliding window within the scheduling cycle at a fixed step size, and the test start time and test station that minimizes the incremental cost are selected; the incremental cost includes incremental time-of-use electricity cost, incremental demand electricity cost, peak limit penalty and waste electricity penalty.

[0023] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. In existing technologies, production scheduling and charge / discharge testing are two independent processes. Production scheduling only focuses on capacity and schedule, while test energy optimization is limited to static pairing within workstations. The lack of temporal linkage between the two leads to random overlap of test loads, persistently high peak power purchases across the plant, and insufficient utilization of the mutual charging / discharging potential among multiple devices. This invention incorporates production cycle constraints, test workstation occupancy constraints, and energy matching rules into a unified scheduling model. Using the production completion time as the earliest start boundary for testing, it achieves sequential linkage between production and testing plans. This reduces idle waiting time after equipment completion and provides a feasible time window for timing matching of multi-device charging / discharging, breaking down the barriers between production resources, testing resources, and energy optimization in the overall architecture.

[0024] 2. Existing technologies generally assume that the charging and discharging path for energy storage cabinet testing is a rigid sequence, with each stage executed continuously and without adjustment; otherwise, testing efficiency and process effectiveness will be affected. This invention breaks through this limitation by setting an adjustable standby buffer time between adjacent charging and discharging stages without changing the total test capacity, charging and discharging rate, or pass / fail criteria. Combined with incremental cost optimization in the decoding process, the discharge window of a single device is dynamically aligned with the peak charging load of the entire plant. This improvement requires no additional hardware investment and does not significantly extend the total test cycle, yet it greatly increases the time overlap of the charging and discharging processes between devices, allowing more discharge energy to be consumed locally by charging equipment within the same plant, significantly reducing grid waste. The improvement in energy utilization efficiency exceeds the expected effect of conventional timing scheduling methods and is non-obvious.

[0025] 3. Existing technologies typically optimize only a single cost dimension, focusing either on saving on time-of-use electricity costs or controlling peak demand. They generally treat the number of production lines and testing stations as fixed conditions, excluding them from the optimization scope, easily leading to idle resources or insufficient configuration. This invention incorporates time-of-use electricity costs, peak demand costs, equipment operating costs, and completion time penalties into a unified objective function. Simultaneously, it uses the number of production lines and testing stations as decision variables for collaborative optimization. By staggering charging and discharging peaks to mitigate overall plant peak demand and by reducing grid electricity purchases through energy redistribution, it dynamically matches the optimal number of production lines and testing stations based on order size, balancing equipment depreciation costs and energy savings, avoiding investment waste caused by over-configuration, and achieving global cost minimization.

[0026] 4. Existing energy optimization solutions often sacrifice cost reduction for extended testing cycles, making it difficult to guarantee order delivery; moreover, most are only suitable for small-batch scenarios with fixed workstation pairings, and the scheduling effect decreases significantly as the scale expands. This invention uses a genetic algorithm combined with sliding window incremental cost optimization, strictly constraining the completion time limit during the optimization process, and compressing equipment waiting time through priority ranking and dynamic workstation allocation; at the same time, the phase adjustment mechanism has a scale effect, the more production and testing equipment there are, the more sufficient the matching space of the charging and discharging window, and the more significant the improvement effect of energy mutual assistance, overcoming the problem of the optimization effect decaying with the scale of traditional methods. It is suitable for small-batch multi-model customized production and can also support large-scale mass production scenarios, with good engineering applicability and scalability. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating the overall process of the method of the present invention. Figure 2 A framework diagram of the coupled scheduling model for production cycle time, testing stations, and energy supply; Figure 3 This is a schematic diagram of the individual encoding structure in a genetic algorithm. Figure 4 Here is the flowchart for decoding the scheduling process; Figure 5 A device-view Gantt chart showing the optimal scheduling scheme for the example; Figure 6 A Gantt chart from the perspective of the test workstation for the optimal scheduling scheme in the example; Figure 7 The graph shows the total net power of the plant versus time-of-use electricity price for the optimal scheduling scheme in this example. Detailed Implementation

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. These embodiments are implemented based on the technical solution of the present invention, providing detailed implementation methods and specific operating procedures; however, the scope of protection of the present invention is not limited to the following embodiments.

[0029] This invention relates to the cross-technical field of energy storage equipment production scheduling and energy management. The core is to achieve coupled scheduling of production cycle, testing station, and energy supply. Based on this, a flexible adjustment mechanism for charge and discharge phases within the testing path is further introduced. Without adding new hardware equipment, changing testing process standards, or altering the total testing capacity, the matching degree of charge and discharge windows and the energy supply rate between equipment are further improved. While ensuring order delivery, the potential for mutual energy support between multiple devices during charge and discharge is deeply explored, reducing the overall plant's electricity purchase cost and peak demand. At the same time, the resource allocation of production lines and testing stations is optimized, balancing equipment investment and electricity costs.

[0030] A collaborative scheduling and management method for energy consumption in the manufacturing, production, and testing of energy storage cabinets. The overall process of this method is as follows: Figure 1 As shown, the process includes four core steps in sequence: parameter acquisition, model construction, algorithm solution, and solution output, as detailed below: S1: Obtain order information and equipment parameters: Obtain order information and full equipment parameters for energy storage cabinets to be manufactured. The equipment parameters specifically include: equipment model, production quantity of each model, and assembly and production time per unit. Rated energy storage capacity Initial state of charge of the equipment And preset optional test charging and discharging paths, the stage division and single stage duration of each test path, and the maximum allowable buffer time. .

[0031] This method pre-sets two standard test charge / discharge paths, which can be flexibly selected according to product testing requirements: Path 1: From the initial Charge to 100% (charging time) ), then discharge to 0% (discharge phase duration) Finally, recharge to 30% (recharge phase duration) ); Path 2: From the initial Discharge to 0% (duration of discharge phase) Recharge to 100% (charging time) ), then discharge to 0% (discharge phase duration) Finally, recharge to 30% (recharge phase duration) ).

[0032] A standby buffer period can be inserted between adjacent charge / discharge phases of each test path, and the buffer duration can be [specified]. Adjustments can be made within a specified range, and the total testing time for a single device must not exceed the preset upper limit. .

[0033] S2: Construct a scheduling model that couples production cycle time, testing stations, and energy supply: The coupled scheduling model framework constructed in this step is as follows: Figure 2 As shown, the model includes four types of constraints and one objective function, where the number of production lines is... With the number of test stations For decision variables to be optimized, joint optimization can be performed within a preset value range.

[0034] The constraints include production cycle time constraints, test station constraints, test path phase constraints, and energy supply constraints.

[0035] Assume the order contains a total of The scheduling cycle for the equipment is [number] units. Minutes, Configuration Parallel production lines and One test station; equipment ( The decision variable is test priority. Test path ( ) and inter-stage buffer time .

[0036] 1. Production cycle time constraints: Each piece of equipment sequentially enters multiple parallel production lines for assembly and production. The completion time of a single piece of equipment is equal to the start time of production plus its own production time, that is: ; At the same time, the same production line can process at most one piece of equipment at any given time; and the testing time of the equipment shall not be earlier than its production completion time, that is, the testing process must wait until production is completed before it can be started.

[0037] 2. Test station constraints: Each device's charge / discharge test occupies one test station throughout the entire process. A single test station can accommodate a maximum of one device at a time, and the test periods for different devices cannot overlap at the same station. The total test time for a single device is the sum of the duration of each stage and the buffer time, and must not exceed [a certain limit]. .

[0038] 3. Test path phase constraint: For the selected test path, the buffer time between adjacent charge / discharge phases satisfies: ; By adjusting The time window for the discharge phase of the device can be shifted, and the matching and optimization with the charging window of other devices can be achieved within the total test duration constraint, thereby improving the charging and discharging efficiency.

[0039] 4. Energy constraints: During the testing phase, each device generates minute-level charge / discharge power curves according to the selected charge / discharge path and buffer duration, covering the entire plant. The net power at a given time is the sum of the power of all devices at that time. ; in For equipment exist The charging and discharging power at any given moment, with charging being positive and discharging being negative.

[0040] The energy matching rule is as follows: within the same minute, the power released by the discharging device shall be given priority to the charging needs of the charging device within the same minute; the portion of the discharge power exceeding the charging needs shall be regarded as waste electricity, and the portion of the charging needs exceeding the discharge power shall be made up by the power purchased from the grid.

[0041] The energy utilization rate is further defined as the core evaluation indicator for the degree of energy mutual assistance and reuse in the quantitative testing process, and the calculation formula is as follows: ; In the formula, This represents the total energy required for charging all test equipment during the scheduling period. This represents the total amount of electricity actually purchased from the grid during the dispatch cycle. A higher energy utilization rate indicates a higher proportion of the equipment's discharge energy being reused locally by charging equipment within the same factory, resulting in less waste electricity in the feeder grid and higher energy utilization efficiency during the testing phase.

[0042] The objective function aims to minimize the sum of grid power purchase costs, equipment operating costs, and completion timeout penalties within the scheduling cycle, i.e.: ; The calculation rules for each cost are as follows: 1. Time-of-use electricity charge: Calculated by accumulating the product of the net power purchased per minute within the dispatch cycle and the corresponding minute-based time-of-use electricity price. ; in For simulating step size, for Time-of-use electricity pricing at any given moment.

[0043] 2. Peak demand charge: This is the product of the peak net purchased power and the demand charge during the dispatching period. ; in This is the demand-based electricity price.

[0044] 3. Equipment operating costs: This is the product of the number of production lines and the operating cost per production line, plus the product of the number of testing stations and the operating cost per testing station. The operating cost per production line is calculated by dividing the production line purchase cost by its useful life, and the operating cost per testing station is calculated by dividing the testing station purchase cost by its useful life. ; in Operating cost per production line Operating cost per test station.

[0045] 4. Completion Timeout Penalty: This is the portion of the scheduling plan's completion time that exceeds the scheduling cycle, multiplied by a penalty coefficient per unit time. ; in The completion time of the scheduling plan. This represents the penalty coefficient for timeout per unit of time.

[0046] S3: Solve the scheduling model using a genetic algorithm: This step uses a genetic algorithm to solve the above coupled scheduling model. The core includes four steps: encoding, decoding, fitness evaluation, and genetic operations. The overall iterative process is as follows: initialize the population → decode and evaluate fitness → selection, crossover, and mutation → determine whether the termination condition is met. If not, return to the iteration; if so, output the historical best individual.

[0047] Individual coding structure such as Figure 3 As shown, an individual consists of four parts: the production line quantity gene, the test station quantity gene, and the first segment. Test priority weights, second paragraph There are 1 test path selection, and the total length of each individual is 1. .in: The production line quantity gene and the test station quantity gene are integers, and each takes a value within a preset range. The first paragraph is Each real-valued priority weight corresponds to the test priority of each device; The second segment consists of N integer path selections, each with a value of 1 or 2, corresponding to the test charging and discharging path selected by each device. Inter-stage buffer time In the decoding stage, the encoding length and algorithm complexity are controlled by determining the encoding through local optimization without occupying encoding bits.

[0048] Decoding process as follows Figure 4 As shown, the processing order of the equipment is first determined in ascending order of priority weight. Then, production and testing stages are arranged for each piece of equipment in sequence. The testing stage simultaneously optimizes the start time, workstation, and stage buffer time. 1. Production process decoding: Select equipment in order of priority and assign it to the earliest available production line. The completion time of equipment production is the sum of the release time of the production line and the equipment production time.

[0049] 2. Testing Phase Decoding: The earliest testable time is the equipment completion time. Starting from this time, a sliding window search is performed within the scheduling cycle at fixed steps to find available time slots for each testing station. For each candidate start time and candidate station, the optimal inter-stage buffer time is searched within the allowed buffer time range. Calculate the incremental cost for each candidate solution, and select the combination of test start time, test station and buffer time that minimizes the incremental cost.

[0050] The incremental cost includes four parts: incremental time-of-use electricity charges, incremental demand electricity charges, peak limit penalties, and waste electricity penalties. Incremental time-of-use electricity cost: The cost of electricity purchased from the grid during the testing period of this equipment; Incremental demand charge: If the peak demand of the entire plant is increased after the equipment is connected, the additional demand charge will be charged. Peak value penalty: When the peak value of the total net power purchase exceeds the preset peak value limit, the penalty is calculated by multiplying the excess amount by a penalty coefficient. Waste power penalty: The penalty corresponding to the portion of the equipment's discharge power that is not offset by the grid and is directly fed back into the grid.

[0051] By optimizing the buffer time locally, the discharge window of the current equipment can be maximized to overlap with the charging window of the already arranged equipment. This improves the overall energy utilization rate and reduces incremental costs without extending the total project duration or adding new hardware.

[0052] The total cost calculated using the objective function is used as the fitness of an individual; the smaller the total cost, the better the individual's fitness.

[0053] Genetic operations include three types: selection, crossover, and mutation, and a historical best individual retention strategy is employed to preserve the optimal solution during the evolutionary process. Selection: A tournament selection method is used, in which a specified number of individuals are randomly selected from the population each time, and the individuals with the best fitness are selected to enter the next generation.

[0054] Crossover: Two-point crossover is used, where two crossover points are randomly selected on the individual coding sequence, and the gene segments between the crossover points of the two parent individuals are exchanged.

[0055] Mutation: Mutation is performed according to the probability of gene position: the mutation method of the gene for the number of production lines and the gene for the number of test stations is to add or subtract 1 from the original value, and to limit the value range of each of them; the mutation of the priority bit is to generate a new random real number; the mutation of the path bit is to switch to another test path.

[0056] S4: Output the optimal scheduling scheme: After the algorithm iteration terminates, the optimal scheduling scheme corresponding to the historical best individual is output, including: the production start and end time, test start and end time, test station number, test path, buffer time between each stage, and the optimal number of production lines and the optimal number of test stations for each device.

[0057] The collaborative scheduling and management system for energy consumption in energy storage cabinet manufacturing, production, and testing implements the above methods and includes four main modules: parameter acquisition module, model building module, optimization solution module, and solution output module.

[0058] The parameter acquisition module is used to acquire order information and equipment parameters, as well as the range of values ​​for the number of production lines and testing stations; the equipment parameters include equipment model, quantity of each model, and production time per unit. Energy storage capacity Initial state of charge Optional test charge / discharge paths, duration of each stage, and maximum allowable buffer duration. .

[0059] The model building module is used to construct a scheduling model that couples production cycle time, test station, and energy supply. Internally, it includes production cycle time constraint units, test station constraint units, test path phase constraint units, energy supply constraint units, and an objective function unit. Production cycle time constraint unit: used to limit each piece of equipment to enter multiple parallel production lines for assembly production in sequence, and the equipment's testing start time is not earlier than its production completion time; Test station constraint unit: used to limit the charge and discharge test of each device to occupy one test station, and the same test station can accommodate a maximum of one device at the same time. Test path phase constraint unit: used to limit the range of buffer duration values ​​between adjacent stages of the test path, as well as the upper limit of the total test duration; Energy counter-constraint unit: used to limit the generation of minute-level charge and discharge power curves for each device during the test phase according to the selected charge and discharge path and buffer duration. The net power of the entire plant is the sum of the power of each device, and the discharge power is preferentially offset against the charging power. Objective function unit: The objective is to minimize the sum of grid power purchase cost, equipment operating cost and completion timeout penalty within the scheduling cycle. The grid power purchase cost includes time-of-use electricity cost and peak demand electricity cost.

[0060] The optimization and solution module is used to solve the scheduling model using a genetic algorithm. Internally, it includes an encoding unit, a decoding unit, a genetic operation unit, and a fitness evaluation unit. Encoding unit: Used to form an individual code based on the number of production lines, the number of test stations, the test priority of each device, and the test path; Decoding unit: Used to prioritize and schedule production and testing for each device, generating a scheduling plan; the execution rules for the testing phase are as follows: the earliest testable time is the time when the device completes production; from that time, the available time slots for each test station are searched in a sliding window within the scheduling cycle at fixed steps; at the same time, the optimal inter-stage buffer time is searched within the allowable range; and the combination of test start time, test station, and buffer time that minimizes incremental cost is selected; the incremental cost includes incremental time-of-use electricity cost, incremental demand electricity cost, peak limit penalty, and waste electricity penalty; Genetic operation units: used to perform selection, crossover, and mutation operations to iteratively evolve the population; Fitness assessment unit: used to calculate the objective function value as the individual's fitness.

[0061] The scheme output module is used to output the optimal scheduling scheme, which includes the production start and end times, test start and end times, test station numbers, test paths, buffer times for each stage, as well as the optimal number of production lines and the optimal number of test stations for each device.

[0062] Specific application examples: Example 1: Production testing and scheduling of small-batch, multi-model energy storage cabinets, as shown in Tables 1-4: In this embodiment, an energy storage equipment production workshop needs to complete the assembly, production, and charge / discharge testing of 12 Type A energy storage cabinets, 12 Type B energy storage cabinets, and 1 Type C energy storage compartment. The workshop is equipped with 10 parallel production lines and 30 testing stations; scheduling cycle... The simulation step size is 1 day (1440 minutes). The test lasted 1 minute; the test was conducted at a 0.5P magnification; each device was initially... It follows a normal distribution with a mean of 30% and a standard deviation of 1%, and is truncated between 27% and 33%; maximum allowable buffer time. Set to 20 minutes.

[0063] Table 1 Equipment Specifications

[0064] Implementing the city's April time-of-use electricity pricing policy, demand-based electricity pricing. The price is 40 yuan / kW, with specific time-of-use electricity prices as follows: Table 2 Time-of-use Electricity Price Table

[0065] After obtaining all parameters according to S1, construct the coupled scheduling model according to S2; initialize a genetic algorithm population with a population size of 145, with each individual having a length of 56 (containing genes for the number of production lines, the number of test stations, and two segments of encoding for each of the 25 devices); decode and evaluate fitness according to S3, and after 300 generations of evolution, use two-point crossover, tournament selection, and gene position mutation, while retaining the historical best individual. The algorithm parameters and equipment cost parameters are as follows: Table 3 Genetic Algorithm Parameter Table

[0066] Table 4 Equipment Cost Parameters

[0067] Finally, the optimal scheduling scheme is output according to S4. To verify the improvement effect of introducing phase adjustment in this invention, the solutions for the "original scheme without phase adjustment" and the "optimized scheme with phase adjustment" are solved and compared respectively.

[0068] In the optimal scheduling scheme obtained in this embodiment, the optimal number of production lines is 3, with 7 vacant; the optimal number of test stations is 13, with 17 vacant; after introducing phase adjustment, the energy utilization rate is significantly improved, the total cost of electricity purchase and the peak demand of the whole plant are significantly reduced, and the completion time of the scheduling scheme is still within the scheduling cycle, with minimal impact on the total project duration.

[0069] The scheduling Gantt chart from the device's perspective is as follows: Figure 5 As shown, the time sequence distribution of production periods, grid charging periods, discharge periods, and auxiliary charging periods for each piece of equipment can be clearly seen, as well as the fine-tuning effect of buffer periods on the discharge window; the scheduling Gantt chart from the perspective of the test station is shown below. Figure 6 As shown, the test paths and timing arrangements of each device at the 13 test stations can be seen; the net power and time-of-use electricity price curves for the entire plant are as follows: Figure 7 As shown, the net power peak is further controlled, and the matching degree of charge and discharge load is significantly improved.

[0070] Example 2: Scheduling verification in large-scale mass production scenarios: To verify the applicability and optimization effect of the method of this invention in large-scale mass production scenarios, simulation verification was carried out for the production testing of 100 Type A energy storage cabinets, 100 Type B energy storage cabinets, and 50 Type C energy storage compartments in a power plant project. The workshop production line configuration ranged from 1 to 10 lines, and the test station configuration ranged from 1 to 30 stations. The scheduling cycle, simulation step size, test multiplier, and initial equipment settings were also considered. The parameters such as time-of-use pricing, demand pricing, and equipment operating cost conversion rules remain consistent with those in Example 1; maximum buffer time. Still set to 20 minutes.

[0071] After optimization using the coupled scheduling method proposed in this invention, the results show that in the production and testing scenario of hundreds of large-scale, multi-model energy storage devices, the method of this invention can still effectively achieve deep collaborative optimization of production cycle time, testing station resources, and energy supply. Moreover, the phase adjustment mechanism can further tap the potential for energy mutual assistance, and the energy supply rate can be further improved in small-batch scenarios, demonstrating a better scale mutual assistance effect. At the same time, it can reasonably configure the number of production lines and testing stations while ensuring order delivery cycle, balancing equipment investment costs and electricity costs, thus verifying the scalability and engineering application value of the method.

[0072] The flexible adjustment mechanism for charge and discharge phases within the test path introduced in this invention solves the technical bias that is common in existing coupled scheduling technologies—that is, the test charge and discharge path is considered to be a fixed process sequence that cannot be adjusted, resulting in low matching degree of charge and discharge windows between equipment, limited improvement of energy utilization rate, and serious waste of discharge energy.

[0073] This mechanism requires no new hardware equipment and does not change the charging and discharging capacity, process standards, or total duration requirements of the energy storage cabinet test. It simply introduces an adjustable standby buffer window between adjacent charging and discharging stages in the test path and performs local optimization in the decoding stage by combining the plant's already scheduled load. This allows the discharge window of a single device to be precisely matched with the peak charging load of the entire plant, significantly improving the energy utilization rate, reducing waste electricity penalties and electricity purchase costs, and further smoothing out the plant's net power fluctuations and reducing peak demand.

[0074] This improvement is fully compatible with the existing production-testing-energy coupled scheduling framework, does not compromise the integrity of the original solution, and brings cost reduction and energy efficiency improvement effects that cannot be predicted by those skilled in the art based on existing technology. It has outstanding substantive features and significant progress.

[0075] In summary, this method significantly reduces grid power purchases and lowers electricity costs by coupling production scheduling with charge / discharge testing and leveraging the mutual support of multiple equipment charging and discharging. It also reduces peak demand across the plant by using incremental cost guidance and peak demand penalties, thereby lowering demand-based electricity costs. Furthermore, it deeply explores the potential for energy mutual support between equipment through flexible adjustment of test path phases. The coordinated arrangement of production and testing processes reduces equipment waiting time and shortens completion time. Simultaneously, it incorporates the number of production lines and testing stations into collaborative optimization, achieving a balance between equipment investment and electricity purchase costs, and minimizing the overall total cost.

[0076] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can still make modifications or equivalent substitutions to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention are within the protection scope of the claims of the present invention.

Claims

1. A method for coordinated scheduling and management of energy consumption in the manufacturing, production, and testing of energy storage cabinets, characterized in that: Includes the following steps: S1: Obtain order information and equipment parameters, including equipment model, quantity of each model, production time of a single unit, energy storage capacity, initial state of charge, and selectable test charging and discharging paths; S2: Construct a scheduling model coupling production cycle time, testing stations, and energy support. This model includes production cycle time constraints, testing station constraints, energy support constraints, and an objective function. The number of production lines and testing stations are decision variables to be optimized. The production cycle time constraint ensures that each piece of equipment sequentially enters multiple parallel production lines for assembly, and the testing start time of each piece of equipment is no earlier than its production completion time. The testing station constraint ensures that each piece of equipment occupies one testing station for charging and discharging testing, and the same testing station can accommodate at most one piece of equipment at any given time. The energy support constraint ensures that during the testing phase, each piece of equipment generates minute-level charging and discharging power curves according to the selected charging and discharging path. The total net power of the plant is the sum of the power of each piece of equipment, with priority given to offsetting charging power support. The objective function minimizes the sum of grid power purchase cost, equipment operating cost, and completion timeout penalty within the scheduling cycle. The grid power purchase cost includes time-of-use electricity charges and peak demand charges. S3: The scheduling model is solved using a genetic algorithm: the number of production lines, the number of test stations, the test priority of each device, and the test path constitute the encoded individuals. During decoding, production and testing are arranged for each device in order of priority, a scheduling scheme is generated, and the objective function value is calculated as the individual fitness. S4: Output the optimal scheduling scheme, which includes the production start and end times, test start and end times, test station numbers, test paths, and the optimal number of production lines and the optimal number of test stations for each device.

2. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 1, characterized in that, The time-of-use electricity charge is calculated by multiplying the net power purchased per minute within the scheduling cycle by the corresponding minute-based time-of-use electricity price; the peak demand charge is the product of the peak net power purchased within the scheduling cycle and the demand price; the equipment operating cost is the product of the number of production lines and the operating cost of a single production line, plus the product of the number of testing stations and the operating cost of a single testing station. The operating cost of a single production line is calculated by dividing the purchase cost of the production line by its service life, and the operating cost of a single testing station is calculated by dividing the purchase cost of the testing station by its service life; the completion timeout penalty is the portion of the scheduling plan's completion time exceeding the scheduling cycle multiplied by a unit time penalty coefficient.

3. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 1, characterized in that, The production process arrangement rule during decoding is as follows: equipment is selected in order of priority and assigned to the earliest available production line. The completion time of equipment production is the sum of the release time of the production line and the equipment production time.

4. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 1, characterized in that, The rules for arranging the testing phase during decoding are as follows: the earliest testable time is the time when the equipment is completed. Starting from that time, the available time slots for each test station are searched in a sliding window within the scheduling cycle at fixed steps, and the test start time and test station that minimizes incremental costs are selected.

5. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 4, characterized in that, The incremental costs include incremental time-of-use electricity charges, incremental demand charges, peak limit penalties, and waste electricity penalties; wherein, the peak limit penalty is calculated by multiplying the excess amount by a penalty coefficient when the peak value of the global net purchased power exceeds the preset peak limit; the waste electricity penalty is the penalty corresponding to the portion of the discharge power that is not offset by the grid and is fed back to the grid.

6. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 1, characterized in that, The specific rules for the priority of discharge power and the offsetting of charging power are as follows: within the same minute, the discharge power of each test device is given priority to supply the charging needs of each test device within that minute. The part of the discharge power exceeding the charging needs is recorded as waste power, and the part of the charging needs exceeding the discharge power is supplemented by electricity purchased from the power grid.

7. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 1, characterized in that, The encoded individual consists of genes representing the number of production lines and the number of test stations, N test priority weights in the first segment, and N test path selections in the second segment, where N is the total number of devices. Genetic operations include two-point crossover, probability mutation based on gene loci, and tournament selection, and a historical best individual retention strategy is adopted. During decoding, the device processing order is determined in ascending order of priority weights. The mutation methods for the production line number gene and the test station number gene are to add or subtract 1 from the original value and limit them to their respective value ranges. The priority bit mutation generates a random real number, and the path bit mutation switches to another test path.

8. The energy consumption collaborative scheduling and management method for energy storage cabinet manufacturing, production, and testing according to claim 1, characterized in that, The test charge and discharge paths include the following two types: Path 1: charge from the initial state of charge to 100%, then discharge to 0%, and then recharge to 30%; Path 2: discharge from the initial state of charge to 0%, then charge to 100%, then discharge to 0%, and then recharge to 30%.

9. A collaborative scheduling and management system for energy consumption in the manufacturing, production, and testing of energy storage cabinets, characterized in that: include: The parameter acquisition module is used to acquire order information and equipment parameters. The equipment parameters include equipment model, quantity of each model, production time of a single unit, energy storage capacity, initial state of charge, and selectable test charging and discharging paths. The model building module is used to construct a scheduling model coupling production cycle time, testing stations, and energy supply constraints. This module includes a production cycle time constraint unit, a testing station constraint unit, an energy supply constraint unit, and an objective function unit. The number of production lines and testing stations are the decision variables to be optimized. The production cycle time constraint unit limits each piece of equipment to sequentially enter multiple parallel production lines for assembly production, and the start time of equipment testing is no earlier than its production completion time. The testing station constraint unit limits each piece of equipment to occupy one testing station for charging and discharging testing, and the same testing station can accommodate at most one piece of equipment at the same time period. The energy supply constraint unit limits each piece of equipment during the testing phase to generate minute-level charging and discharging power curves according to the selected charging and discharging path. The total net power of the plant is the sum of the power of each piece of equipment, with discharging power preferentially offsetting charging power. The objective function unit aims to minimize the sum of grid power purchase cost, equipment operating cost, and completion timeout penalty within the scheduling cycle. The grid power purchase cost includes time-of-use electricity charges and peak demand charges. An optimization solution module is used to solve the scheduling model using a genetic algorithm. The optimization solution module includes an encoding unit, a decoding unit, a genetic operation unit, and a fitness evaluation unit. The encoding unit is used to construct encoded individuals based on the number of production lines, the number of test stations, the test priority of each device, and the test path. The decoding unit is used to arrange production and testing for each device in order of priority to generate a scheduling scheme. The fitness evaluation unit is used to calculate the objective function value as the individual fitness. The scheme output module is used to output the optimal scheduling scheme, which includes the production start and end times, test start and end times, test station numbers, test paths, and the optimal number of production lines and the optimal number of test stations for each device.

10. The energy storage cabinet manufacturing, production, testing, and energy consumption collaborative scheduling management system according to claim 9, characterized in that, The decoding unit executes the test scheduling rules as follows: the earliest testable time is the equipment production completion time. From that time, the unit searches for available time periods for each test station in a sliding window within the scheduling cycle at a fixed step size, and selects the test start time and test station that minimizes incremental costs. The incremental costs include incremental time-of-use electricity costs, incremental demand electricity costs, peak limit penalties, and waste electricity penalties.

Citation Information

Patent Citations

  • PCS testing system, methods, equipment, and storage media for energy storage containers

    CN116365671B

  • Test method and test platform for twin trawling test of energy storage battery prefabricated cabin

    CN119199621A

  • Battery pack multi-production line collaborative scheduling method, storage medium and battery production system

    CN119784099A

  • Lithium battery industry MES system production scheduling method based on deep learning

    CN120087729A