An industrial adjustable load response cost optimization system and method

Abstract

The application discloses an industrial adjustable load response cost optimization system, which comprises a task acquisition and analysis module, an adjustable load screening module, an individual dynamic cost calculation module and a collaborative optimization decision module. The task acquisition and analysis module acquires a demand response task and analyzes target response capacity and total response time length. The adjustable load screening module screens a candidate load set from a pre-built industrial adjustable load characteristic library based on the total response time length. The individual dynamic cost calculation module calculates individual dynamic response costs of each device in the candidate load set based on operation data of each device in the candidate load set through a trained response cost coupling model. The collaborative optimization decision module establishes an optimization decision model with the minimum total cost as a target and a penalty term constructed in combination with collaborative operation constraints among devices, and obtains an optimal response device combination and optimal response depth of each device through solution. The application effectively reduces user-side response costs under the premise of meeting power grid regulation requirements.

Description

Technical Field

[0001] This invention relates to the field of power demand response technology, specifically to an industrial adjustable load response cost optimization system and method. Background Technology

[0002] With the development of virtual power plant technology, industrial adjustable loads, as an important demand response resource, have made optimizing and controlling their response costs crucial for improving the economics of virtual power plants. Existing technologies typically construct an industrial adjustable load characteristic library by assigning static tags to industrial electrical equipment. Based on these tags, a set of equipment capable of participating in demand response is selected, and the combination of equipment participating in the response is determined by price signals or a single economic indicator. This approach treats the response cost of equipment as a static value or a simple linear variable, failing to fully consider the dynamic impact of industrial load on response costs due to response duration, response depth, and real-time equipment operating status. It also ignores the amplifying effect between production disturbances and equipment lifespan degradation, resulting in insufficient accuracy in cost calculations, optimization schemes deviating from actual production constraints, and difficulty in achieving optimal total response cost with the determined combination of response equipment. Summary of the Invention

[0003] The purpose of this invention is to provide an industrial adjustable load response cost optimization system and method. This invention achieves refined modeling and collaborative optimization of the total cost of industrial adjustable load response, effectively reducing user-side response costs while meeting the grid regulation requirements.

[0004] To achieve this objective, the present invention provides an industrial adjustable load response cost optimization system, comprising: The task acquisition and parsing module is used to acquire the demand response tasks for power grid load regulation, parse the demand response tasks, and obtain the target response capacity and total response time required for the response. The adjustable load screening module is used to select a set of candidate loads from a pre-built industrial adjustable load characteristic library based on the total response time; The individual dynamic cost calculation module is used to calculate the individual dynamic response cost of the corresponding target electrical equipment based on the equipment operation-related data of each target electrical equipment in the candidate load set and through a trained response cost coupling model. The collaborative optimization decision module is used to determine the combination of response devices that meets the target response capacity from the candidate load set. The objective function is to minimize the sum of the individual dynamic response costs corresponding to the combination of response devices. The module also incorporates a penalty term based on the collaborative operation constraints between the target electrical devices to establish an optimization decision model. The optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination of response devices are obtained through the optimization decision model.

[0005] Preferably, the specific process of obtaining the demand response task for power grid load regulation, parsing the demand response task, and obtaining the target response capacity and total response time required for the response is as follows: The system acquires demand response tasks for power grid load regulation. The demand response task instructions are transmitted in the form of data packets conforming to a specific communication protocol. The system calls a task parsing subroutine to decode and parse the data packets of the demand response task and extract the key technical parameters of the demand response task from the structured data fields. The key technical parameters include target response capacity and total response time. The target response capacity refers to the total electrical power value that the aggregated load needs to adjust within a specified time period. The total response time refers to the duration from the start of the response to the end of the response.

[0006] Preferably, the specific process of selecting a candidate load set from a pre-built industrial adjustable load characteristic library based on the total response time is as follows: The pre-built industrial adjustable load characteristic library is a database that includes various electrical devices and configures multi-dimensional tags for each electrical device. The multi-dimensional tags include a response speed tag that quantifies the time required for the application electrical device to start executing power adjustment from load adjustment command, an adjustable depth tag that identifies the maximum adjustable power range of the application electrical device, and a device state constraint tag that defines the process and safe operation boundary of the application electrical device. Based on the total response time, the industrial adjustable load characteristic library is traversed, and the response delay time indicated by the response speed tag of each electrical device is used to filter out electrical devices whose response delay time does not exceed the total response time, thus obtaining a candidate load set.

[0007] Preferably, the specific process for training the response cost coupling model is as follows: Collect historical datasets for a preset time period. The historical datasets include the event occurrence time and actual total cost of each load adjustment event, the actual response depth and actual response duration of each target electrical device, a snapshot of the real-time operating status of each target electrical device when the corresponding load adjustment event occurs, and the cumulative number of adjustments made by each target electrical device within the preset historical period. The actual total cost includes the actual production loss value and the actual equipment life depreciation cost. An initial global model is constructed. Based on the actual response depth, actual response time, and economic value corresponding to the unit time capacity of each target electrical device, the direct production loss cost component of each target electrical device is calculated. The calculation formula is as follows: in, The direct production loss cost component for the target electrical equipment. The actual response depth of the target electrical equipment. The actual response time of the target electrical equipment. The economic value corresponding to the unit time production capacity of the target electrical equipment; Based on the actual response depth, actual response duration, and cumulative adjustment times of each target electrical device within a preset historical period, the equipment life depreciation cost component of each target electrical device is calculated. The direct production loss cost component and the equipment life depreciation cost component of each target electrical equipment are nonlinearly coupled, and the expression is: in, The individual dynamic response cost of the target electrical equipment. The cost component of the equipment's lifespan depreciation for the target electrical equipment. The weighting coefficient for the direct production loss cost component. The weighting coefficient for the equipment lifespan depreciation cost component. The coupling coefficient is... It is a nonlinear coupling term; The coupling coefficients in the global model were optimized using historical datasets and optimization algorithms. Weighting coefficients for direct production loss cost components Weighting coefficients for equipment lifespan depreciation costs Joint training and optimization are performed to minimize the error between the predicted individual dynamic response cost TC of the target electrical equipment and the actual total cost of the corresponding target electrical equipment in the historical dataset, thereby obtaining a well-trained response cost coupling model.

[0008] Preferably, the specific process for calculating the equipment life depreciation cost component of each target electrical device based on the actual response depth, actual response duration, and cumulative adjustment count within a preset historical period is as follows: Based on the actual response depth of each target electrical device, calculate the ratio of the actual response depth of the target electrical device to the rated maximum adjustment depth of the device. The calculation formula is as follows: in, The ratio of the actual response depth of the target electrical equipment to the rated maximum adjustment depth of the equipment. The actual response depth of the target electrical equipment. The rated maximum adjustment depth of the target electrical equipment; Based on the actual response time of each target electrical device, calculate the ratio of the actual response time of the target electrical device to the longest allowable single adjustment time of the device. The calculation formula is as follows: in, This is the ratio of the actual response time of the target electrical equipment to the longest allowable single adjustment time of the equipment. The actual response time of the target electrical equipment. The maximum allowable single adjustment time for the target electrical equipment; Regarding the depth ratio The aforementioned time ratio The multidimensional fatigue accumulation factor of the target electrical equipment is obtained by weighting the cumulative number of adjustments made by the target electrical equipment within a preset historical period. The calculation formula is as follows: in, The multidimensional fatigue accumulation factor for the target electrical equipment. The cumulative number of adjustments made by the target electrical equipment within a preset historical period. Depth ratio The preset weighting coefficients, The ratio of duration The preset weighting coefficients, For the cumulative number of adjustments The preset weighting coefficients; Based on the multidimensional fatigue accumulation factor of the target electrical equipment The corresponding benchmark single-adjustment depreciation value is used to calculate the equipment life depreciation cost component of the target electrical equipment. The calculation formula is as follows: in, The base single-adjustment depreciation value for the target electrical equipment.

[0009] Preferably, the specific process of calculating the individual dynamic response cost of the corresponding target electrical equipment based on the equipment operation-related data of each target electrical equipment in the candidate load set, using a trained response cost coupling model, is as follows: The equipment operation-related data of the target electrical equipment includes the total response time in the demand response task, the maximum adjustable power range determined based on the adjustable depth label of the target electrical equipment, and equipment operation status data; After obtaining the candidate load set, the equipment operation-related data of each target electrical device in the candidate load set are input into the trained response cost coupling model, and the response depth of the target electrical device is used as the optimization variable. The optimization variable takes values ​​within the maximum adjustable power range of the target electrical device. Based on the optimized variables, the direct production loss cost component and equipment life depreciation cost component of the target electrical equipment are calculated through the trained response cost coupling model, thereby obtaining the individual dynamic response cost of the target electrical equipment.

[0010] Preferably, the specific process of determining the combination of response devices that meets the target response capacity from the candidate load set, using the minimization of the sum of the individual dynamic response costs corresponding to the combination of response devices as the objective function, and combining it with a penalty term constructed based on the cooperative operation constraints between the target electrical devices, to establish an optimization decision model is as follows: The cooperative operation constraints include synchronous start-stop constraints, mutually exclusive operation constraints, and capacity coupling constraints between devices; based on the physical connection and process flow relationship between each target electrical device, a cooperative constraint matrix is ​​constructed to characterize the cooperative operation constraints between the target electrical devices; From the candidate load set, a combination of response devices that satisfies the target response capacity is determined. The objective function is established to minimize the sum of the individual dynamic response costs corresponding to each of the response device combinations. The target response capacity serves as an equality constraint, and the collaborative constraint matrix serves as an inequality constraint. The severity of violations of the inequality constraints is quantified as a penalty term and added to the objective function to form a fitness function. This fitness function measures the merits of each response device combination, thus completing the construction of the optimization decision model. The fitness function expression is: in, The value of the fitness function. The total number of target electrical devices in the candidate load set; Indicates the first i The gating state variables of a target electrical device It can take the value 0 or 1; Indicates the first i Response depth of each target electrical device; Indicates the first i The response depth of the target electrical equipment is The corresponding individual dynamic response cost at that time; This is a penalty item.

[0011] Preferably, the specific process of obtaining the optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination of response devices by optimizing the decision model is as follows: The optimization decision model is solved through iterative optimization to obtain the optimal response device combination that minimizes the sum of individual dynamic response costs corresponding to the response device combination under the premise of satisfying the target response capacity and collaborative operation constraints. The optimal response depth corresponding to each target power device in the optimal response device combination is obtained through the fitness function.

[0012] A method for optimizing the cost of industrial adjustable load response includes the following steps: Obtain the demand response tasks for power grid load regulation, analyze the demand response tasks, and obtain the target response capacity and total response time required for the response; Based on the total response time, a candidate load set is obtained by screening from a pre-built library of industrial adjustable load characteristics; Based on the equipment operation-related data of each target electrical device in the candidate load set, the individual dynamic response cost of the corresponding target electrical device is calculated through the trained response cost coupling model. The combination of response devices that meets the target response capacity is determined from the candidate load set. The objective function is to minimize the sum of the individual dynamic response costs corresponding to the combination of response devices. A penalty term based on the cooperative operation constraints between target electrical devices is combined to establish an optimization decision model. The optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination are obtained through the optimization decision model.

[0013] A computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described method.

[0014] The beneficial effects of this invention are: This invention uses multi-dimensional labels based on response speed, adjustable depth, and equipment status constraints to screen candidate loads, retaining only electrical equipment that meets response timeliness and process safety requirements. This avoids unfeasible response schemes from the source, ensuring that selected electrical equipment has real-time response capabilities and safe operating conditions. The invention constructs a response cost coupling model, nonlinearly coupling direct production losses and equipment lifespan depreciation. It quantifies the dynamic impact of response depth, response duration, and historical adjustment frequency on costs, restoring the mutual amplification effect of production disturbances and lifespan decay, thus solving the problem of large deviations in traditional static cost calculations. With the goal of minimizing total response cost, this invention incorporates collaborative operation constraint penalties to construct an optimization decision model. During the solution process, it automatically avoids on-site constraints such as synchronous start-stop, mutually exclusive operation, and capacity coupling, ensuring that the optimal response equipment combination is both economical and feasible, effectively reducing user-side response costs. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 This is a flowchart of the present invention. Detailed Implementation

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1 An industrial adjustable load response cost optimization system, such as Figure 1 As shown, it includes: The task acquisition and parsing module is used to acquire demand response tasks for power grid load regulation, parse the demand response tasks, and obtain the target response capacity (target response capacity refers to the total amount of power consumption that needs to be reduced or increased according to the demand response task) and the total response duration (total response duration refers to the length of time from the start time of the response specified by the demand response task to the end time of the response). This design can quickly acquire and accurately parse demand response tasks, extract the core parameters of target response capacity and total response duration, and provide a quantitative basis for subsequent load screening, cost calculation and optimization decision-making. The adjustable load screening module is used to screen a set of candidate loads from a pre-built industrial adjustable load characteristic library based on the total response time. This design screens the set of candidate loads based on the total response time and the multi-dimensional tags of the industrial adjustable load characteristic library, retaining only the electrical equipment whose response delay time meets the requirements and eliminating the electrical equipment that does not meet the requirements, thus ensuring that the candidate electrical equipment has basic response capabilities. The individual dynamic cost calculation module is used to calculate the individual dynamic response cost of the corresponding target electrical equipment based on the equipment operation-related data of each target electrical equipment in the candidate load set and through a trained response cost coupling model. This design, by combining the trained response cost coupling model with equipment operation-related data to calculate the individual dynamic response cost, breaks through the limitations of static cost calculation and realizes dynamic and differentiated measurement of response cost. The collaborative optimization decision module is used to determine the combination of response equipment that meets the target response capacity from the candidate load set. The objective function is to minimize the sum of the individual dynamic response costs corresponding to the combination of response equipment. In combination with the penalty term constructed based on the collaborative operation constraints between the target electrical equipment, an optimization decision model is established. The optimal response equipment combination and the optimal response depth corresponding to each target electrical equipment in the optimal response equipment combination are obtained through the optimization decision model. (The optimal response depth refers to the specific power regulation amount (usually in kW or MW) used by the target electrical equipment to perform the demand response task, representing how much power consumption the corresponding target electrical equipment needs to reduce or increase.) This design, by constructing an optimization decision model and solving for the optimal response equipment combination and optimal response depth, not only meets the grid target response capacity requirements but also complies with the collaborative operation constraints between the target electrical equipment, and can achieve optimal response cost while meeting the grid regulation requirements.

[0017] Regarding the cooperative operation constraints among the target electrical devices, some optimized technical solutions include: Cooperative operation constraints describe the relationships of cooperation and mutual restriction among multiple selected target electrical devices, representing the association rules that the selected target electrical devices must follow when working collaboratively. For example: Device A and Device C must start and stop simultaneously because they share the same pipeline (synchronization constraint); Devices D and E cannot operate simultaneously (mutual exclusion constraint); the total power of Devices G, H, and I is limited by the capacity of the same transformer (capacity coupling constraint).

[0018] In the above technical solution, the specific process of obtaining the demand response task for power grid load regulation, parsing the demand response task, and obtaining the target response capacity and total response time required for the response is as follows: Through the communication interface deployed in the virtual power plant control center, edge computing node or industrial user-side server, the system receives demand response task instructions issued by the power grid dispatching agency, power trading platform or virtual power plant aggregator in real time. The demand response task instructions are transmitted in the form of data packets that conform to specific communication protocols (such as IEC 60870-5-104, IEC 61850 or DL / T634.5104). The system calls the task parsing subroutine to decode and parse the data packets of the demand response task, and extracts the key technical parameters of the demand response task from the structured data fields. The key technical parameters include the target response capacity and the total response time. The target response capacity refers to the total amount of power that the demand response task requires to be reduced or increased. The total response time refers to the duration from the start time to the end time of the response as specified in the demand response task. For example, the parsed demand response task information may be "reduce the load by 500kW between 14:00 and 15:00", then the target response capacity is 500kW and the total response time is 1 hour. The parsed target response capacity and total response time are stored in the task cache area of ​​the virtual power plant control center, edge computing nodes, or industrial user-side servers. The above design obtains the target response capacity and total response time by standardizing the parsing process of demand response tasks and using specific communication protocol decoding and structured field extraction methods, providing a unified and reliable parameter basis for subsequent adjustable load screening and cost calculation.

[0019] In the above technical solution, the specific process of selecting a candidate load set from a pre-built industrial adjustable load characteristic library based on the total response time is as follows: The pre-built industrial adjustable load characteristic library is a database that includes various electrical equipment and configures multi-dimensional tags for each electrical equipment (such as electric arc furnace, rolling mill, water pump, and fan). The multi-dimensional tags include a response speed tag that quantifies the time required for the application equipment to start executing power adjustment from load adjustment command, an adjustable depth tag that identifies the maximum adjustable power range of the application equipment, and an equipment state constraint tag that defines the process and safe operation boundary of the application equipment. Based on the total response time, the industrial adjustable load characteristic library is traversed. According to the response delay time indicated by the response speed tag of each electrical device (the response delay time is the time from when the electrical device receives the instruction to when it starts to act), electrical devices whose response delay time does not exceed the total response time are selected from the industrial adjustable load characteristic library. For example, if the total response time is 30 minutes, electrical devices with a response delay time exceeding 30 minutes will be directly excluded. As long as the response delay time of the electrical device is less than or equal to the total response time of the demand response task, the electrical device can start in time and continue to work within the total response time of the demand response task, thus obtaining the electrical device that meets the requirements. The system checks whether the real-time operating parameters (such as current temperature, pressure, speed, material state, etc.) of the electrical equipment that meets the requirements satisfy the process safety boundary conditions defined by the equipment status constraint label. Only electrical equipment whose current operating state fully complies with its own safety constraints is judged to have real-time availability for immediate response. Electrical equipment that does not meet the current safety constraints (such as equipment under maintenance, in critical process stages, or with parameters exceeding limits) will be eliminated. The electrical equipment that ultimately meets both the task time requirements and its own real-time safety conditions constitutes the candidate load set for the aforementioned demand response task. The above design is based on multi-dimensional labels (response speed label, adjustable depth label, equipment status constraint label) of the industrial adjustable load characteristic library. It uses the response delay time not exceeding the total response time as the screening condition to accurately screen the candidate load set, realize the pre-verification of the response capability of electrical equipment, ensure that the selected electrical equipment can perform power adjustment within the total response time, and at the same time rely on the equipment status constraint label to ensure the process safety of the response process, thereby improving the effectiveness and availability of candidate loads.

[0020] Regarding the equipment status constraint label, some optimized technical solutions include: the equipment status constraint label is used to describe the real-time operating limitations of a single electrical device, representing the physical or technological boundary conditions under which the electrical device is allowed to participate in regulation at a certain moment, and determining whether the electrical device is eligible to participate in the response at that time. For example: whether the current operating temperature of the electrical device allows for load reduction, whether the material is in a critical process stage, and whether the safety protection device is in a critical state, etc.

[0021] In the above technical solution, the specific process of training the response cost coupling model is as follows: Historical datasets (which can be set to the past year) are collected from the energy management systems, production execution systems, and equipment maintenance systems of the target industrial users for a preset time period. The historical datasets include the event occurrence time and actual total cost of each load adjustment event, the actual response depth and actual response duration of each target electrical device, a snapshot of the real-time operating status of each target electrical device when the corresponding load adjustment event occurs, and the cumulative number of adjustments made by each target electrical device within the preset historical period. The actual total cost includes the actual production loss value and the actual equipment life depreciation cost. An initial global model is constructed. Based on the actual response depth, actual response duration, and economic value (yuan / hour) of the corresponding target electrical equipment's unit time (e.g., hourly) capacity, the direct production loss cost component of each target electrical equipment is calculated. The calculation formula is as follows: in, The direct production loss cost component for the target electrical equipment. The actual response depth of the target electrical equipment. The actual response time of the target electrical equipment. The economic value corresponding to the unit time production capacity of the target electrical equipment; Based on the actual response depth (e.g., 0.2, representing 20%), actual response time, and cumulative adjustment times within a preset historical period, the equipment life depreciation cost component of each target electrical device is calculated. The direct production loss cost component and the equipment life depreciation cost component of each target electrical equipment are nonlinearly coupled, and the expression is: in, The individual dynamic response cost of the target electrical equipment. The cost component of the equipment's lifespan depreciation for the target electrical equipment. The weighting coefficient for the direct production loss cost component. The weighting coefficient for the equipment lifespan depreciation cost component. The coupling coefficient is used to quantify the nonlinear mutual amplification effect; This is a nonlinear coupling term used to simulate the mutually reinforcing effect between production disturbances and equipment lifespan degradation in real industrial production. For example, when equipment condition deteriorates due to lifespan reduction, the same power adjustment may lead to more severe production losses (such as increased scrap rate), and the resulting additional costs are reflected through the nonlinear coupling term. The coupling coefficients in the global model were optimized using historical datasets and optimization algorithms. Weighting coefficients for direct production loss cost components Weighting coefficients for equipment lifespan depreciation costs Joint training and optimization are performed to minimize the error (e.g., root mean square error) between the predicted individual dynamic response cost (TC) of the target electrical equipment and the actual total cost of the corresponding target electrical equipment in the historical dataset. This results in a well-trained response cost coupling model. After training, the response cost coupling model is validated using a historical dataset that was not used in the training to evaluate its prediction accuracy. When the prediction error is less than a preset threshold (the preset threshold can be set according to the requirements for cost prediction accuracy in actual applications, the quality of historical data, and the expected performance of the response cost coupling model; the preset threshold can be set to 15%), the response cost coupling model is considered complete and can be used for real-time calculation of individual dynamic response costs. The above design trains the response cost coupling model using a historical dataset, nonlinearly coupling direct production loss costs and equipment lifespan depreciation costs to simulate the mutual amplification effect of production disturbances and equipment lifespan decay. By optimizing the weight coefficients and coupling coefficients through optimization algorithms, the error between the predicted individual dynamic response cost of the target electrical equipment and the actual total cost of the target electrical equipment is minimized, thereby improving the calculation accuracy of individual dynamic response costs.

[0022] In the above technical solution, the specific process for calculating the equipment life depreciation cost component of each target electrical device based on the actual response depth, actual response duration, and cumulative adjustment count within a preset historical period is as follows: Based on the actual response depth of each target electrical device, calculate the ratio of the actual response depth of the target electrical device to the rated maximum adjustment depth of the device. The calculation formula is as follows: in, The ratio of the actual response depth of the target electrical equipment to the rated maximum adjustment depth of the equipment. The actual response depth of the target electrical equipment. The rated maximum adjustment depth of the target electrical equipment; Based on the actual response time of each target electrical device, calculate the ratio of the actual response time of the target electrical device to the longest allowable single adjustment time of the device. The calculation formula is as follows: in, This is the ratio of the actual response time of the target electrical equipment to the longest allowable single adjustment time of the equipment. The actual response time of the target electrical equipment. The maximum allowable single adjustment time for the target electrical equipment; Maximum allowable single adjustment time for the target electrical equipment Some optimized technical solutions include: the maximum allowable adjustment time for a single adjustment of the target electrical equipment refers to the longest time that the target electrical equipment can maintain a certain adjustment state without causing excessive damage. For example, the maximum allowable adjustment time for a single adjustment of a certain electrical equipment is 2 hours. If it exceeds 2 hours, it needs to be switched back to the rated state for restoration.

[0023] Regarding the depth ratio The aforementioned time ratio The multidimensional fatigue accumulation factor of the target electrical equipment is obtained by weighting the cumulative number of adjustments made by the target electrical equipment within a preset historical period. The calculation formula is as follows: in, The multidimensional fatigue accumulation factor for the target electrical equipment. The cumulative number of adjustments made by the target electrical equipment within a preset historical period. Depth ratio The preset weighting coefficients reflect the degree to which the response depth contributes to the reduction of equipment lifespan; The ratio of duration The preset weighting coefficients reflect the degree to which response time contributes to the reduction of equipment lifespan; For the cumulative number of adjustments The preset weighting coefficient reflects the contribution of historical adjustment times to equipment lifespan reduction, and the cumulative adjustment times. Preset weighting coefficients As the duration of the adjustment event decreases, the closer the adjustment event is to the current time, the greater its impact on the equipment's lifespan. Based on the multidimensional fatigue accumulation factor of the target electrical equipment The corresponding benchmark single-adjustment depreciation value is used to calculate the equipment life depreciation cost component of the target electrical equipment. The calculation formula is as follows: in, The above design uses the baseline single-adjustment depreciation value of the target electrical equipment as a reference. It calculates a multi-dimensional fatigue accumulation factor by weighting the depth ratio, duration ratio, and cumulative adjustment number. Combined with the baseline single-adjustment depreciation value, it accurately calculates the equipment life depreciation cost component, quantifies the comprehensive impact of response depth, response duration, and historical adjustment frequency on the life of the target electrical equipment, and reflects the dynamic characteristics of fatigue accumulation of the target electrical equipment, making the calculation of equipment life depreciation cost more consistent with the physical loss law of the target electrical equipment.

[0024] Regarding the benchmark single-adjustment depreciation value of the target electrical equipment, some optimized technical solutions include: the benchmark single-adjustment depreciation value of the target electrical equipment can be predetermined based on the original asset value, design service life, and the number of rated adjustments allowed within the design service life. For example, if the original asset value of the target electrical equipment is 1 million yuan and the number of rated adjustments allowed within its service life is 10,000, then the benchmark single-adjustment depreciation value of the target electrical equipment can be roughly estimated as 100 yuan / adjustment.

[0025] In the above technical solution, the specific process of calculating the individual dynamic response cost of the corresponding target electrical equipment based on the equipment operation-related data of each target electrical equipment in the candidate load set and through a trained response cost coupling model is as follows: The equipment operation-related data of the target electrical equipment includes the total response time in the demand response task, the maximum adjustable power range determined based on the adjustable depth label of the target electrical equipment, and equipment operation status data; After obtaining the candidate load set, the equipment operation-related data of each target electrical device in the candidate load set is input into the trained response cost coupling model. The response depth of the target electrical device is used as the optimization variable. The optimization variable is taken within the maximum adjustable power range of the target electrical device (the final value of the optimization variable is the optimal response depth). Based on the aforementioned optimization variables, the direct production loss cost component and equipment lifespan depreciation cost component of the target electrical equipment are calculated using a trained response cost coupling model, thereby obtaining the individual dynamic response cost of the target electrical equipment. The above design uses response depth as an optimization variable, combines total response time, maximum adjustable power range, and equipment operating status data, and calculates the individual dynamic response cost through the response cost coupling model to achieve the linkage measurement of individual dynamic response cost and response depth, so as to adapt to the response cost variation pattern under different response depths.

[0026] In the above technical solution, the specific process of determining the combination of response devices that meets the target response capacity from the candidate load set, minimizing the sum of the individual dynamic response costs corresponding to the combination of response devices as the objective function, and combining it with a penalty term constructed based on the cooperative operation constraints between the target electrical devices to establish an optimization decision model is as follows: The cooperative operation constraints include synchronous start-stop constraints, mutually exclusive operation constraints, and capacity coupling constraints between devices; based on the physical connection and process flow relationship between each target electrical device, a cooperative constraint matrix is ​​constructed to characterize the cooperative operation constraints between the target electrical devices; For the aforementioned collaborative constraint matrix, some optimized technical solutions include: collaborative constraint matrix It is The matrix, where The total number of target electrical devices in the candidate load set; elements in the matrix. Used to characterize the i The target electrical equipment and the first j The relationship between the target electrical equipment; From the candidate load set, determine the combination of response equipment that satisfies the target response capacity (i.e., the total power adjusted by all target electrical equipment in the combination can meet the target response capacity). With the objective of minimizing the sum of the individual dynamic response costs corresponding to the combination of response equipment, establish an objective function. Use the target response capacity as an equality constraint and the collaborative constraint matrix as an inequality constraint. Quantify the severity of violating the inequality constraints as a penalty term and add it to the objective function to form a fitness function. The fitness function measures the merits of each combination of response equipment, thereby completing the construction of the optimization decision model (such as a mixed-integer nonlinear programming model). The fitness function expression is: in, The fitness function is used to measure the quality of a response plan (i.e., a combination of response devices). The smaller the value, the lower the total response cost of the corresponding response plan (the total response cost is the sum of the individual dynamic response costs corresponding to the combination of response devices) and the higher the feasibility of the response plan. The total number of target electrical devices in the candidate load set; Indicates the first i The gating state variables of a target electrical device It can take the value 0 or 1. =1 indicates the first i One target electrical device is selected in the response device combination; =0 indicates the first i The target electrical equipment was not selected and is not in the response equipment combination; Indicates the first i Response depth of each target electrical device; Indicates the first i The response depth of the target electrical equipment is The corresponding individual dynamic response cost at that time; As a penalty item, The magnitude of the value is positively correlated with the severity of the violation of the cooperative operation constraints by the currently evaluated response scheme; the above design constructs a cooperative constraint matrix based on the cooperative operation constraints of the equipment, integrates the penalty term into the objective function to form a fitness function, and automatically avoids constraint violations during the optimization process, which can ensure the optimality and feasibility of the response scheme. Regarding the aforementioned penalty item Some optimization techniques include: penalty terms. The form can be expressed as: in, This is the penalty coefficient; Indicates the first u The constraint function of the cooperative operation constraint; This represents the set of gating state variables for all target electrical devices. q represents the set of response depths of all target electrical devices; q represents the total number of cooperative operation constraints.

[0027] In the above technical solution, the specific process of obtaining the optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination of response devices by optimizing the decision model is as follows: Based on the synchronous start-stop constraints, mutually exclusive operation constraints, and capacity coupling constraints between the devices, a collaborative constraint matrix is ​​constructed to characterize the collaborative operation constraints between the target electrical devices. From the candidate load set, determine the combination of response devices that satisfies the target response capacity. With the objective of minimizing the sum of the individual dynamic response costs corresponding to each of the response device combinations, establish an objective function. Use the target response capacity as an equality constraint and the collaborative constraint matrix as an inequality constraint. Quantify the severity of inequality violations as a penalty term and add it to the objective function to form a fitness function. The fitness function measures the merits of each response device combination, thus completing the construction of the mixed-integer nonlinear programming model. The fitness function expression is: in, The fitness function value is used to measure the quality of a response plan. The smaller the value, the lower the total response cost of the corresponding response plan and the higher the feasibility of the response plan; The total number of target electrical devices in the candidate load set; Indicates the first i The gating state variables of a target electrical device It can take the value 0 or 1. =1 indicates the first i One target electrical device is selected in the response device combination; =0 indicates the first i The target electrical equipment was not selected and is not in the response equipment combination; Indicates the first i Response depth of each target electrical device; Indicates the first i The response depth of the target electrical equipment is The corresponding individual dynamic response cost at that time; As a penalty item, The magnitude of this value is positively correlated with the severity of the violation of cooperative operation constraints by the currently evaluated response scheme; The mixed-integer nonlinear programming model is solved through iterative optimization, searching the solution space of the decision variables. Response schemes with high fitness values ​​(i.e., high total response cost or serious constraint violation) are continuously evaluated and eliminated, while better response schemes are retained and evolved until convergence is met. This yields the optimal response device combination that minimizes the sum of individual dynamic response costs corresponding to the response device combination, while satisfying the target response capacity and collaborative operation constraints. The optimal response depth for each target electrical device in the optimal response device combination is obtained through the fitness function. This design, through iterative optimization of the decision model, locks the optimal response device combination with the minimum total response cost and the optimal response depth for each target electrical device, while satisfying the target response capacity and collaborative operation constraints. This ensures accurate output of the optimization results and guarantees the economy and executability of the response scheme.

[0028] In some preferred embodiments of the present invention, after determining the optimal combination of response devices, if the optimal combination of response devices includes target electrical devices with different response speed label levels, dynamic power allocation is performed to further optimize the response cost in the time dimension. Specific implementation steps include: The target electrical equipment in the optimal response equipment combination is divided into a first response speed level group (fast response group) and a second response speed level group (slow response group) according to the level of the response speed label. The target electrical equipment in the fast response group has a faster response speed than that in the slow response group. Within the total response duration T, the response power undertaken by the first response speed level group and the second response speed level group is dynamically allocated according to a preset power allocation function. The power allocation function takes time t1 as the independent variable and must satisfy the core constraint: at any time t1 within the total response duration T, the sum of the response power undertaken by the devices in the first response speed level group and the second response speed level group is always equal to the target response capacity. P target The expression is: P fast (t1) + P slow (t1) = P target in, P target For the target response capacity, P fast(t1) represents the response power of the first response speed group that decreases with time. P slow (t1) represents the response power of the second response speed group as it increases over time; This allows for a smooth transfer of power from high-cost, high-speed equipment to low-cost, slow-speed equipment. Depending on the different dynamic characteristics of the target electrical equipment, this allocation can be achieved using two typical functional forms: one is a piecewise linear function, in which the first response speed group undertakes the entire target response capacity at the initial moment of the response. P target The second response speed level group is 0; the ramp time required from the initial moment until the second response speed level group reaches its rated power. T ramp Within, the response power of the first response speed level group is from P target The response power of the second response speed group linearly increases from 0 to 0, decreasing linearly to 0. P target The absolute values ​​of the decreasing and increasing slopes are equal; during the climbing time... T ramp During the period from the end of the response to the end of the total response duration T, the second response speed level group independently assumes the entire target response capacity. P target The response power of the first response speed group remains at 0, with the ramp time... T ramp Pre-determined based on the response speed label of the target electrical equipment in the second response speed level group; The second is the exponential function combination, where the first response speed level group undertakes the entire target response capacity at the initial moment of the response. P target Within the total response time T, the response power of the first response speed group decays exponentially, and the expression for the exponential decay function is: in, This represents the response power undertaken by the first response speed level group at time t1. This represents the decay time constant of the first response speed level group; e It is a natural constant; The response power of the second response speed group increases exponentially, and the exponential growth function expression is: in, This indicates the response power undertaken by the second response speed group at time t1. This represents the decay time constant of the first response speed level group; The combination of the exponential decay function and the exponential growth function can satisfy the power conservation constraint at any time. After solving, the optimal response power curve that changes with time is generated. The optimal response power curve is used to guide the fine-grained power scheduling of the first response speed level group and the second response speed level group throughout the entire response period.

[0029] Example 2 A method for optimizing the response cost of industrial adjustable loads, such as Figure 2 As shown, the demand response task is obtained and the target response capacity and total response time are parsed out; based on the total response time, a candidate load set is selected from the pre-built industrial adjustable load characteristic library; based on the operating data of each device in the candidate load set, the individual dynamic response cost of each device is calculated through the trained response cost coupling model; with the goal of minimizing the total cost and combined with the penalty term constructed by the coordination constraints between devices, an optimization decision model is established to solve for the optimal combination of response devices and the optimal response depth of each device.

[0030] The specific methods for optimizing adjustable load response costs include the following steps: Obtain the demand response tasks for power grid load regulation, analyze the demand response tasks, and obtain the target response capacity and total response time required for the response; Based on the total response time, a candidate load set is obtained by screening from a pre-built library of industrial adjustable load characteristics; Based on the equipment operation-related data of each target electrical device in the candidate load set, the individual dynamic response cost of the corresponding target electrical device is calculated through the trained response cost coupling model. The combination of response devices that meets the target response capacity is determined from the candidate load set. The objective function is to minimize the sum of the individual dynamic response costs corresponding to the combination of response devices. A penalty term based on the cooperative operation constraints between target electrical devices is combined to establish an optimization decision model. The optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination are obtained through the optimization decision model.

[0031] Example 3 A computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in Embodiment 2.

[0032] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0033] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0034] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0035] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0036] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims of the invention.

[0037] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. An industrial adjustable load response cost optimization system, characterized in that, It includes: The task acquisition and parsing module is used to acquire the demand response tasks for power grid load regulation, parse the demand response tasks, and obtain the target response capacity and total response time required for the response. The adjustable load screening module is used to select a set of candidate loads from a pre-built industrial adjustable load characteristic library based on the total response time; The individual dynamic cost calculation module is used to calculate the individual dynamic response cost of the corresponding target electrical equipment based on the equipment operation-related data of each target electrical equipment in the candidate load set and through a trained response cost coupling model. The collaborative optimization decision module is used to determine the combination of response devices that meets the target response capacity from the candidate load set. The objective function is to minimize the sum of the individual dynamic response costs corresponding to the combination of response devices. The module also incorporates a penalty term based on the collaborative operation constraints between the target electrical devices to establish an optimization decision model. The optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination of response devices are obtained through the optimization decision model.

2. The industrial adjustable load response cost optimization system according to claim 1, characterized in that: The specific process of obtaining the demand response task for power grid load regulation, parsing the demand response task, and obtaining the target response capacity and total response time required for the response is as follows: The system acquires demand response tasks for power grid load regulation. The demand response task instructions are transmitted in the form of data packets conforming to a specific communication protocol. The system calls a task parsing subroutine to decode and parse the data packets of the demand response task and extract the key technical parameters of the demand response task from the structured data fields. The key technical parameters include target response capacity and total response time. The target response capacity refers to the total electrical power value that the aggregated load needs to adjust within a specified time period. The total response time refers to the duration from the start of the response to the end of the response.

3. The industrial adjustable load response cost optimization system according to claim 1, characterized in that: Based on the total response time, the specific process of selecting a candidate load set from a pre-built library of industrial adjustable load characteristics is as follows: The pre-built industrial adjustable load characteristic library is a database that includes various electrical devices and configures multi-dimensional tags for each electrical device. The multi-dimensional tags include a response speed tag that quantifies the time required for the application electrical device to start executing power adjustment from load adjustment command, an adjustable depth tag that identifies the maximum adjustable power range of the application electrical device, and a device state constraint tag that defines the process and safe operation boundary of the application electrical device. Based on the total response time, the industrial adjustable load characteristic library is traversed, and the response delay time indicated by the response speed tag of each electrical device is used to filter out electrical devices whose response delay time does not exceed the total response time, thus obtaining a candidate load set.

4. The industrial adjustable load response cost optimization system according to claim 1, characterized in that: The specific process for training the response cost coupling model is as follows: Collect historical datasets for a preset time period. The historical datasets include the event occurrence time and actual total cost of each load adjustment event, the actual response depth and actual response duration of each target electrical device, a snapshot of the real-time operating status of each target electrical device when the corresponding load adjustment event occurs, and the cumulative number of adjustments made by each target electrical device within the preset historical period. The actual total cost includes the actual production loss value and the actual equipment life depreciation cost. An initial global model is constructed. Based on the actual response depth, actual response time, and economic value corresponding to the unit time capacity of each target electrical device, the direct production loss cost component of each target electrical device is calculated. The calculation formula is as follows: in, The direct production loss cost component for the target electrical equipment. The actual response depth of the target electrical equipment. The actual response time of the target electrical equipment. The economic value corresponding to the unit time production capacity of the target electrical equipment; Based on the actual response depth, actual response duration, and cumulative adjustment times of each target electrical device within a preset historical period, the equipment life depreciation cost component of each target electrical device is calculated. The direct production loss cost component and the equipment life depreciation cost component of each target electrical equipment are nonlinearly coupled, and the expression is: in, The individual dynamic response cost of the target electrical equipment. The cost component of the equipment's lifespan depreciation for the target electrical equipment. The weighting coefficient for the direct production loss cost component. The weighting coefficient for the equipment lifespan depreciation cost component. The coupling coefficient is... It is a nonlinear coupling term; The coupling coefficients in the global model were optimized using historical datasets and optimization algorithms. Weighting coefficients for direct production loss cost components Weighting coefficients for equipment lifespan depreciation costs Joint training and optimization are performed to minimize the error between the predicted individual dynamic response cost TC of the target electrical equipment and the actual total cost of the corresponding target electrical equipment in the historical dataset, thereby obtaining a well-trained response cost coupling model.

5. The industrial adjustable load response cost optimization system according to claim 4, characterized in that: The specific process for calculating the equipment life depreciation cost component of each target electrical device based on its actual response depth, actual response duration, and cumulative adjustment count within a preset historical period is as follows: Based on the actual response depth of each target electrical device, calculate the ratio of the actual response depth of the target electrical device to the rated maximum adjustment depth of the device. The calculation formula is as follows: in, The ratio of the actual response depth of the target electrical equipment to the rated maximum adjustment depth of the equipment. The actual response depth of the target electrical equipment. The rated maximum adjustment depth of the target electrical equipment; Based on the actual response time of each target electrical device, calculate the ratio of the actual response time of the target electrical device to the longest allowable single adjustment time of the device. The calculation formula is as follows: in, This is the ratio of the actual response time of the target electrical equipment to the longest allowable single adjustment time of the equipment. The actual response time of the target electrical equipment. The maximum allowable single adjustment time for the target electrical equipment; Regarding the depth ratio The aforementioned time ratio The multidimensional fatigue accumulation factor of the target electrical equipment is obtained by weighting the cumulative number of adjustments made by the target electrical equipment within a preset historical period. The calculation formula is as follows: in, The multidimensional fatigue accumulation factor for the target electrical equipment. The cumulative number of adjustments made by the target electrical equipment within a preset historical period. Depth ratio The preset weighting coefficients, The ratio of duration The preset weighting coefficients, For the cumulative number of adjustments The preset weighting coefficients; Based on the multidimensional fatigue accumulation factor of the target electrical equipment The corresponding benchmark single-adjustment depreciation value is used to calculate the equipment life depreciation cost component of the target electrical equipment. The calculation formula is as follows: in, The base single-adjustment depreciation value for the target electrical equipment.

6. An industrial adjustable load response cost optimization system according to claim 3 or 5, characterized in that: Based on the equipment operation-related data of each target electrical device in the candidate load set, the specific process of calculating the individual dynamic response cost of the corresponding target electrical device through a trained response cost coupling model is as follows: The equipment operation-related data of the target electrical equipment includes the total response time in the demand response task, the maximum adjustable power range determined based on the adjustable depth label of the target electrical equipment, and equipment operation status data; After obtaining the candidate load set, the equipment operation-related data of each target electrical device in the candidate load set are input into the trained response cost coupling model, and the response depth of the target electrical device is used as the optimization variable. The optimization variable takes values ​​within the maximum adjustable power range of the target electrical device. Based on the optimized variables, the direct production loss cost component and equipment life depreciation cost component of the target electrical equipment are calculated through the trained response cost coupling model, thereby obtaining the individual dynamic response cost of the target electrical equipment.

7. The industrial adjustable load response cost optimization system according to claim 6, characterized in that: The specific process of determining the combination of response devices that meets the target response capacity from the candidate load set, minimizing the sum of the individual dynamic response costs corresponding to the combination of response devices as the objective function, and combining it with a penalty term constructed based on the cooperative operation constraints between the target electrical devices, to establish an optimization decision model is as follows: The cooperative operation constraints include synchronous start-stop constraints, mutually exclusive operation constraints, and capacity coupling constraints between devices; based on the physical connection and process flow relationship between each target electrical device, a cooperative constraint matrix is ​​constructed to characterize the cooperative operation constraints between the target electrical devices; From the candidate load set, a combination of response devices that satisfies the target response capacity is determined. The objective function is established to minimize the sum of the individual dynamic response costs corresponding to each of the response device combinations. The target response capacity serves as an equality constraint, and the collaborative constraint matrix serves as an inequality constraint. The severity of violations of the inequality constraints is quantified as a penalty term and added to the objective function to form a fitness function. This fitness function measures the merits of each response device combination, thus completing the construction of the optimization decision model. The fitness function expression is: in, The value of the fitness function. The total number of target electrical devices in the candidate load set; Indicates the first i The gating state variables of a target electrical device It can take the value 0 or 1; Indicates the first i Response depth of each target electrical device; Indicates the first i The response depth of the target electrical equipment is The corresponding individual dynamic response cost at that time; This is a penalty item.

8. The industrial adjustable load response cost optimization system according to claim 7, characterized in that: The specific process of obtaining the optimal combination of response devices and the optimal response depth for each target electrical device in the optimal combination of response devices by optimizing the decision model is as follows: The optimization decision model is solved through iterative optimization to obtain the optimal response device combination that minimizes the sum of individual dynamic response costs corresponding to the response device combination under the premise of satisfying the target response capacity and collaborative operation constraints. The optimal response depth corresponding to each target power device in the optimal response device combination is obtained through the fitness function.

9. A method for optimizing the cost of industrial adjustable load response, characterized in that, It includes the following steps: Obtain the demand response tasks for power grid load regulation, analyze the demand response tasks, and obtain the target response capacity and total response time required for the response; Based on the total response time, a candidate load set is obtained by screening from a pre-built library of industrial adjustable load characteristics; Based on the equipment operation-related data of each target electrical device in the candidate load set, the individual dynamic response cost of the corresponding target electrical device is calculated through the trained response cost coupling model. The combination of response devices that meets the target response capacity is determined from the candidate load set. The objective function is to minimize the sum of the individual dynamic response costs corresponding to the combination of response devices. A penalty term based on the cooperative operation constraints between target electrical devices is combined to establish an optimization decision model. The optimal combination of response devices and the optimal response depth corresponding to each target electrical device in the optimal combination are obtained through the optimization decision model.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method of claim 9.

Images