Deviation assessment-considered industrial park demand response decision-making method

Abstract

The invention discloses an industrial park demand response decision-making method considering deviation assessment, and relates to the technical field of industrial park demand response decision-making, and the method comprises the steps: obtaining the power grid instruction, electricity price, load and new energy output prediction data of an industrial park in a future demand response period, and the operation parameters of energy storage equipment, interruptible load equipment and the like; on the basis of the obtained data, an optimization decision model with the purpose of reducing the comprehensive operation cost is constructed, the model quantifies demand response subsidy in a segmented mode according to the deviation interval of the actual response amount and the target amount, and linear expression of a segmented revenue structure is achieved through logic constraints; in the response execution process, rolling optimization is started according to the time step length, generated actual operation data are called at each optimization node, the current accumulated deviation is evaluated, the influence of the deviation on the availability of subsequent adjustable resources is predicted in combination with an energy storage energy change rule and a system adjustment capability evolution trend, and a deviation influence prediction result is output; and based on a deviation influence prediction result, dynamically adjusting a risk control strategy of the decision model at the current optimization node, and generating a correction control scheme.

Description

Technical Field

[0001] This invention relates to the field of industrial park demand response decision-making technology, and in particular to a decision-making method for industrial park demand response that takes deviation assessment into account. Background Technology

[0002] Industrial park demand response decision-making technology refers to a comprehensive set of modeling, optimization, and control methods used in power systems to proactively adjust electricity consumption behavior in response to grid dispatching requirements and to obtain economic subsidies, targeting industrial parks as typical load aggregators. This is achieved by optimizing the scheduling of their internal adjustable resources and responding to peak shaving or valley filling commands from the power grid. Therefore, improving the intelligence and security of industrial park demand response decision-making through advanced technologies is one of the most pressing issues to be addressed.

[0003] In the field of demand response decision-making in industrial parks, traditional optimization models often employ linearization assumptions or simplifications, which fail to accurately express such nonlinear revenue structures. This leads to distorted calculations of subsidy income and deviations of optimization decisions from the true optimal solution. Furthermore, most models only constrain the cumulative response amount, neglecting the continuity of time, which can easily result in discontinuous response behavior. Although the total amount meets the target, the discontinuity can lead to invalid responses and penalties. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a decision-making method for industrial park demand response that considers deviation assessment. This addresses the problem that traditional optimization models often employ linearization assumptions or simplifications, which fail to accurately express such nonlinear revenue structures, leading to distorted subsidy income calculations and deviations of optimization decisions from the true optimal solution. Furthermore, most models only constrain the cumulative response amount, neglecting temporal continuity, which can easily result in discontinuous response behavior. Although the total amount meets the target, the discontinuity can lead to invalid responses and penalties.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a decision-making method for industrial park demand response that considers deviation assessment, comprising: Acquire forecast data on power grid commands, electricity prices, load and renewable energy output of industrial parks during future demand response cycles, as well as operating data of energy storage and interruptible load equipment, and construct a set of adjustable resource units; Based on the set of adjustable resource units, an optimization decision-making model is constructed with the goal of reducing overall operating costs. The optimization decision-making model quantifies the demand response subsidy in segments according to the deviation range between the actual response amount and the target amount, and realizes the linear expression of the segmented revenue structure through logical constraints. During the response execution process, rolling optimization is initiated according to the time step. At each optimization node, the actual operating data that has occurred is called to assess the current cumulative deviation. Combining the energy storage energy change pattern and the evolution trend of system regulation capability, the impact of deviation on the availability of subsequent adjustable resources is predicted, and the deviation impact prediction result is output. Based on the prediction results of the deviation impact, the risk control strategy of the decision model is dynamically adjusted at the current optimization node to generate a corrected control scheme. The modified control scheme is input into a digital twin simulation environment synchronized with the physical system. Multi-condition disturbance tests are conducted using a model that includes the power grid topology and equipment dynamic characteristics to verify the stability and boundary compliance of the scheme under abnormal scenarios. After the response is executed, based on the actual output curve and the planned output curve of the adjustable resource unit set, the marginal contribution attribution method is used to quantify the causal contribution of the adjustable resource unit set to the response deviation, and the deviation responsibility weight of the adjustable resource unit set is output. Based on the deviation responsibility weight, the assessment loss ratio is allocated to the corresponding adjustable resource units, generating a unit-level performance evaluation report, and updating and optimizing the initial value of the risk control strategy in the decision-making model.

[0007] As a preferred embodiment of the industrial park demand response decision-making method considering deviation assessment described in this invention, the specific steps for obtaining predicted data on grid commands, electricity prices, load and renewable energy output of the industrial park during the future demand response cycle, as well as operating data of energy storage and interruptible load equipment, and constructing an adjustable resource unit set are as follows: The peak shaving or valley filling command signals for each time period within the future response time are obtained from the power grid dispatching platform. These command signals are binary variables. When the... The value is 1 when peak-shaving response is required, and 0 otherwise, denoted as . ; The value is 1 when a valley-filling response is required, and 0 otherwise. ; Obtain time-of-use electricity prices for future periods from the electricity market system. ; Based on historical electricity consumption data and weather factors, the baseline power of the industrial park under no-response conditions is calculated. ; The photovoltaic power generation forecast values ​​for different time periods are obtained through the photovoltaic power plant monitoring system. ; Obtain the rated capacity, maximum charge / discharge power, and charging efficiency of the energy storage system. Discharge efficiency Upper and lower limits of state of charge and and initial state of charge ; The adjustable load set, initial power, response power, maximum number of calls per day, and set of allowed response periods are obtained from the interruptible load control system and used to establish the start-stop logic variables, cumulative call count constraints, and time period feasibility constraints for the interruptible load in the optimization model.

[0008] As a preferred embodiment of the industrial park demand response decision-making method considering deviation assessment as described in this invention, the following steps are taken: Based on a set of adjustable resource units, an optimization decision-making model is constructed with the goal of reducing overall operating costs. This optimization decision-making model quantifies demand response subsidies in segments according to the deviation range between the actual response amount and the target amount, and achieves a linearized expression of the segmented revenue structure through logical constraints. The specific steps are as follows: The objective function is set to minimize the overall operating cost, and its expression is: ; in, for Electricity price at any time for Total power of the industrial park at any time To reduce peak response subsidies, To fill the valley response subsidy income; Define actual power consumption The compositional relationship is expressed as: ; in, This represents the actual peak reduction response. This represents the actual valley-filling response. This represents the deviation of power from the baseline during non-demand response periods; Introducing constraints to limit the range of the response, the expression is: in, This is the upper limit coefficient for response capability. and Let the variables be 0-1, representing whether peak shaving or valley filling is needed, and satisfying the following conditions: in, It is a very large positive number used to force the response behavior to occur only when the command is issued.

[0009] As a preferred embodiment of the industrial park demand response decision-making method considering deviation assessment described in this invention, the following steps are taken: During the response execution process, rolling optimization is initiated according to time steps. At each optimization node, actual operating data that has occurred is retrieved to assess the current cumulative deviation. Furthermore, by combining the energy storage energy change pattern with the evolution trend of system regulation capability, the impact of the deviation on the availability of subsequent adjustable resources is predicted, and the deviation impact prediction result is output. The specific steps are as follows: The rolling optimization cycle is set to 15 minutes, at each optimization trigger time. Read from the start of the response until Actual charging and discharging power of energy storage at different times and ; Calculate the current actual state of charge based on the energy conservation principle of energy storage. Its recursive formula is: in, For time intervals; Compare Compared with the original optimization plan The deviation is calculated using the following expression: ; according to Calculate the maximum discharge capacity and maximum charging capacity that energy storage can provide during the remaining time period; By combining the remaining number of interruptible load calls with the available time period, assess whether the system's overall resilience in subsequent time periods meets the remaining response target.

[0010] As a preferred embodiment of the industrial park demand response decision-making method considering deviation assessment as described in this invention, the specific steps of dynamically adjusting the risk control strategy of the decision model at the current optimization node based on the deviation impact prediction results to generate a modified control scheme are as follows: When the prediction results indicate that the subsequent adjustment capacity is sufficient, the original objective function is maintained unchanged; When the prediction results show that the adjustment capability is limited, a bias penalty term is introduced into the objective function, and the adjusted objective function is: in, For the first Response deviation at any moment This is a penalty weighting coefficient that increases over time, used to enhance the suppression of bias in later stages; The optimization model is re-solved to generate a new energy storage charging and discharging plan and an interruptible load scheduling sequence. The revised control scheme is output to update the control instructions for periods that have not been executed.

[0011] As a preferred embodiment of the industrial park demand response decision-making method considering deviation assessment described in this invention, the following steps are taken: Before issuing the corrective control scheme, the corrective control scheme is input into a digital twin simulation environment synchronized with the physical system. Multi-condition disturbance tests are conducted using a model that includes the power grid topology and equipment dynamic characteristics to verify the stability and boundary compliance of the scheme under abnormal scenarios. Establish a topology model of the park's power distribution network in the digital twin system, including transformers, feeder impedance, load nodes, and the location of distributed power source access; Configure a dynamic response model for the energy storage system to simulate its charge / discharge switching delay and efficiency degradation characteristics; The modified control scheme is loaded into the simulation model as an input command; Perturbations are applied during the simulation, including scenarios such as sudden drop in photovoltaic output, interruptible load failure, and communication delay. Monitor tie-line power, node voltage, and response trajectory during the simulation process to determine whether the operating boundary constraints are met.

[0012] As a preferred embodiment of the industrial park demand response decision-making method considering deviation assessment as described in this invention, the following steps are taken after the response is executed: based on the actual output curve and planned output curve of the adjustable resource unit set, the marginal contribution attribution method is used to quantify the causal contribution of the adjustable resource unit set to the response deviation, and the deviation responsibility weight of the adjustable resource unit set is output. The actual and planned power of energy storage, interruptible loads, and distributed photovoltaic power during each time period within the response cycle; Using all resource combinations as the sample space, calculate the total response of the system under different subset combinations; Based on the Shapley value formula, the marginal response increment of each resource in all possible combination sequences is averaged to obtain its fair contribution to the total response. The overall response deviation is allocated according to the proportion of each resource contribution value to determine the deviation share that each unit should bear.

[0013] 8. The decision-making method for industrial park demand response considering deviation assessment as described in claim 7, characterized in that: the specific steps of allocating the assessment loss proportion to the corresponding adjustable resource units based on deviation responsibility weights, generating a unit-level performance evaluation report, and updating and optimizing the initial value of the risk control strategy in the decision-making model are as follows: The total amount of subsidies deducted due to failure to meet the response standards will be allocated to each responsible unit according to the calculated impact weights. For devices that bear a higher deviation weight, their scheduling priority will be reduced in the next round of response tasks; By calling upon the full-process simulation data stored in the digital twin system, the key time periods and dominant factors in which deviations occurred can be analyzed; The deviation correction coefficient in the baseline load forecasting model is revised, and the growth function shape of the risk penalty weight in the rolling optimization is adjusted.

[0014] In a second aspect, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the decision-making method for industrial park demand response considering deviation assessment as described in the first aspect of the present invention.

[0015] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the decision-making method for industrial park demand response considering deviation assessment as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By constructing a segmented subsidy quantification model that considers deviation assessment, the demand response benefit structure is refined and linearly solved. Combined with a rolling optimization mechanism and multi-resource collaborative scheduling such as energy storage and interruptible loads, the economy and robustness of the response process are effectively improved. The introduction of digital twin simulation verification and multi-condition disturbance testing enhances the stability and security of the control scheme under abnormal scenarios. The responsibility decomposition mechanism based on marginal contribution attribution realizes the fair quantification of deviation responsibility and the reasonable allocation of assessment losses. Through feedback, the prediction accuracy and control strategy are continuously optimized, thereby improving the compliance, autonomous decision-making ability and overall operational efficiency of industrial parks participating in demand response. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the decision-making method for industrial park demand response that considers deviation assessment in Example 1. Detailed Implementation

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] Example, refer to Figure 1 This embodiment of the invention provides a decision-making method for industrial park demand response that considers deviation assessment, comprising the following steps: S1. Obtain forecast data on power grid commands, electricity prices, load and new energy output of the industrial park during the future demand response cycle, as well as the operation data of energy storage and interruptible load equipment, and construct a set of adjustable resource units; Furthermore, peak shaving or valley filling command signals for each moment within the future response period are obtained from the power grid dispatching platform. These command signals are binary variables. When the... The value is 1 when peak-shaving response is required, and 0 otherwise, denoted as . ; The value is 1 when a valley-filling response is required, and 0 otherwise. ; Obtain time-of-use electricity prices for future periods from the electricity market system. ; Based on historical electricity consumption data and weather factors, the baseline power of the industrial park under no-response conditions is calculated. ; The photovoltaic power generation forecast values ​​for different time periods are obtained through the photovoltaic power plant monitoring system. ; Obtain the rated capacity, maximum charge / discharge power, and charging efficiency of the energy storage system. Discharge efficiency Upper and lower limits of state of charge and and initial state of charge ; The adjustable load set, initial power, response power, maximum number of calls per day, and set of allowable response periods are obtained from the interruptible load control system and used to establish the start-stop logic variables, cumulative call count constraints, and time period feasibility constraints of the interruptible load in the optimization model. It should be noted that by obtaining binary peak shaving / valley filling command signals from the power grid dispatching platform, accurate identification and logical isolation of response task types are achieved, avoiding false responses. Combined with time-of-use pricing, baseline power, and new energy output forecasts, a high-precision economic evaluation model can be constructed. At the same time, comprehensive collection of operating parameters of energy storage and interruptible loads provides a refined modeling foundation for subsequent multi-resource collaborative optimization, improving the practicality and executability of the decision-making model.

[0023] S2. Based on the set of adjustable resource units, construct an optimization decision model with the goal of reducing overall operating costs. The optimization decision model quantifies the demand response subsidy in segments according to the deviation range between the actual response amount and the target amount, and realizes the linear expression of the segmented revenue structure through logical constraints. Furthermore, let's define the objective function as minimizing the overall operating cost, expressed as: ; in, for Electricity price at any time for Total power of the industrial park at any time To reduce peak response subsidies, To fill the valley response subsidy income; Define actual power consumption The compositional relationship is expressed as: ; in, This represents the actual peak reduction response. This represents the actual valley-filling response. This represents the deviation of power from the baseline during non-demand response periods; Introducing constraints to limit the range of the response, the expression is: in, This is the upper limit coefficient for response capability. and Let the variables be 0-1, representing whether peak shaving or valley filling is needed, and satisfying the following conditions: Setting peak shaving response subsidy income For the actual response quantity The piecewise function is expressed as follows: in, For the first Real-time peak reduction response. For the first Continuously reduce the peak target response volume. Price subsidies for peak-shaving response units , , In response to the target compliance rate threshold, , This is a discount factor used to reflect the severity of rewards or penalties for insufficient or excessive responses. Introducing auxiliary continuous variables This indicates that the actual response falls into the first... For a portion of an interval, the following condition is met: Define the subsidy income corresponding to each interval. And introduce 0-1 integer variables Indicates the first Whether an interval is activated, satisfying: Set a mutual exclusion constraint for each interval to ensure that only one interval is active at a time: Set upper and lower bound constraints for each interval to ensure that the variable values ​​take the correct range: in, It is a very large positive number, used to force the response behavior to occur only when the command is issued; Similarly, the calculation method for the grain filling subsidy can be obtained. when During the valley filling period, we have: ; Where M is a very large positive number, which can be set to 1,000,000; when During non-valley filling periods, there are =0; Demand response rate constraint: ; ; Also includes: Within the demand response period, constraints are imposed on the effective response time. The minimum cumulative effective response time constraint is achieved through the following mathematical expression: in, For the first A binary variable indicating whether a peak-shaving effective response occurs at a given time. As an auxiliary 0-1 variable, used to represent the first... The peak-shaving response falls at the first moment in the segmented subsidy model. The activation state of each interval For the first A binary variable indicating whether a valid valley-filling response has occurred at time 1. The value is 1 if a valid valley-filling response exists at that time, and 0 otherwise. As an auxiliary 0-1 variable, used to represent the first... The valley-filling response falls on the first step in the segmented subsidy model. The activation state of each interval; For the entire demand response period, The preset minimum cumulative peak-shaving effective response time, This is the preset minimum cumulative effective response time for valley filling; The optimization model also includes upper and lower power limits, expressed as: ; in, For the first The total power output of the virtual power plant at any given time. and The first The minimum and maximum power allowed by the communication line at all times. This is the set of all scheduled time periods; The optimization model also includes power balance constraints, expressed as follows: ; in, For the first Total load power of the park at any time For the first The interruptible load is in the first Response power at any given time For interruptible load sets, For the first An external input power source (such as the power grid) in the first... Power supply at any time For external input power supply collection, For the first The electric vehicle in the first The charging or discharging power at any given time. For electric vehicles For the first The photovoltaic power generation unit in the first Actual output at any moment A collection of photovoltaic equipment. For the first The energy storage system in the first Output power at any moment It is a collection of energy storage systems.

[0024] It should be noted that by transforming the nonlinear piecewise subsidy function into a linearized structure composed of auxiliary variables, and introducing interval mutual exclusion constraints and upper and lower bound constraints, the originally unsolvable nonlinear revenue model is successfully transformed into a standard mixed-integer linear programming problem. The global optimal solution can be obtained within an acceptable engineering time. At the same time, setting effective response time constraints ensures that the response continuity meets the grid assessment requirements, while power upper and lower bound constraints and power balance constraints guarantee the safety of system operation and energy conservation. The overall modeling method takes into account economy, compliance and computability.

[0025] S3. During the response execution process, rolling optimization is initiated according to the time step. At each optimization node, the actual operating data that has occurred is called to assess the current cumulative deviation. Combined with the energy storage energy change law and the evolution trend of system regulation capability, the impact of deviation on the availability of subsequent adjustable resources is predicted, and the deviation impact prediction result is output. Furthermore, the rolling optimization cycle is set to 15 minutes, at each optimization trigger moment. Read from the start of the response until Actual charging and discharging power of energy storage at different times and ; Calculate the current actual stored energy based on the energy conservation law. Its recursive formula is: ; in, For time intervals, , These are the charging and discharging efficiencies, respectively. Based on rated energy capacity Calculate the current actual state of charge. : ; Compare Compared with the original optimization plan Calculate the deviation : ; The deviation This reflects the energy execution error of the energy storage system during the initial response process. Based on the current SOC and energy storage characteristics, the maximum discharge capacity and maximum charging capacity that energy storage can provide during the remaining time period are estimated as follows: ; The calculation results, combined with the remaining number of interruptible load calls, available time periods, and response power, comprehensively assess whether the system's overall adjustment capability in subsequent periods is sufficient to meet the remaining response target.

[0026] It should be noted that by setting a 15-minute rolling optimization cycle and reading the actual charging and discharging power of the energy storage in real time, and combining this with the recursive formula for the state of charge to dynamically assess the current energy storage state deviation, the remaining regulation capacity can be accurately predicted. This mechanism effectively overcomes the problem of misjudgment of resource availability caused by the accumulation of prediction errors, improves the system's ability to perceive future response potential, provides a reliable basis for proactive risk control, and mitigates local execution deviations. Transforming into global resource availability prediction, it achieves a leap from state monitoring to capacity prediction. Under the influence of uncertainties such as fluctuations in new energy output and load forecasting deviations, it can identify the risk of early depletion of energy storage or lack of discharge capacity in the later stages, avoid overall response failure due to resource misjudgment, and improve the robustness and foresight of the decision-making system.

[0027] S4. Based on the prediction results of the deviation impact, dynamically adjust the risk control strategy of the decision model at the current optimization node and generate a corrected control scheme. Furthermore, when the prediction results indicate sufficient subsequent adjustment capability, the original objective function is maintained unchanged; When the prediction results show that the adjustment capability is limited, a bias penalty term is introduced into the objective function, and the adjusted objective function is: in, For the first Response deviation at any moment This is a penalty weighting coefficient that increases over time, used to enhance the suppression of bias in later stages; Indicates the first The actual response deviation at time t is defined as: ; That is, the difference between the target response and the actual response; The penalty weight coefficient increases over time, and its value satisfies < +1, and designed as a linear or exponential growth function, such as: ; in, Total response time Basic penalty coefficient; The optimization model is re-solved to generate a new energy storage charging and discharging plan and an interruptible load scheduling sequence. The revised control scheme is output to update the control instructions for periods that have not been executed. It should be noted that by introducing a deviation penalty term that increases over time, the risk control strategy is dynamically evolved: a certain deviation is allowed in the early stage of the response to retain adjustment margin, and the suppression is strengthened in the later stage to prevent deviation from the target. This avoids the imbalance problem of over-adjustment in the early stage and no resources available in the later stage. The adaptive penalty mechanism realizes the dynamic evolution of the risk control strategy, enabling the optimization model to have intelligent adjustment capabilities of perception-evaluation-response. It solves the imbalance problem of over-adjustment in the early stage and no resources available in the later stage or conservative in the early stage and out of control in the later stage in the traditional static model, and improves the robustness, economy and compliance of the demand response process.

[0028] S5. Input the modified control scheme into a digital twin simulation environment synchronized with the physical system, and use a model that includes the power grid topology and equipment dynamic characteristics to conduct multi-condition disturbance tests to verify the stability and boundary compliance of the scheme under abnormal scenarios. Furthermore, a topology model of the park's power distribution network is established in the digital twin system, including transformers, feeder impedance, load nodes, and the location of distributed power source access. Configure a dynamic response model for the energy storage system to simulate its charge / discharge switching delay and efficiency degradation characteristics; The modified control scheme is loaded into the simulation model as an input command; Perturbations are applied during the simulation, including scenarios such as sudden drop in photovoltaic output, interruptible load failure, and communication delay. Monitor tie-line power, node voltage, and response trajectory during the simulation process to determine whether the operating boundary constraints are met. It should be noted that by establishing a simulation model in the digital twin environment that includes the grid topology and equipment dynamic characteristics, and applying typical disturbance conditions such as sudden drop in photovoltaic output, interruptible load failure, and communication delay, the stability and boundary compliance of the modified control scheme can be fully verified in the virtual space. This pre-verification mechanism effectively avoids the risk of exceeding limits caused by model simplification or unforeseen disturbances in actual control, and improves the safety of the scheme and the feasibility of engineering implementation.

[0029] S6. After the response is executed, based on the actual output curve and the planned output curve of the adjustable resource unit set, the marginal contribution attribution method is used to quantify the causal contribution of the adjustable resource unit set to the response deviation, and the deviation responsibility weight of the adjustable resource unit set is output. Furthermore, the actual and planned power of energy storage, interruptible loads, and distributed photovoltaic power during each period of the response cycle; Using all resource combinations as the sample space, calculate the total response of the system under different subset combinations; Based on the Shapley value formula, the marginal response increment of each resource in all possible combination sequences is averaged to obtain its fair contribution to the total response. The overall response deviation is allocated according to the proportion of each resource contribution value to determine the deviation share that each unit should bear; It should be noted that by collecting actual and planned output data of each resource, and fairly allocating the marginal response increment of each unit under all resource combinations based on the Shapley value, the quantitative attribution of deviation responsibility is realized. This method overcomes the unfairness of traditional responsibility allocation based on capacity or number of calls, and can accurately identify key responsible units with insufficient response or failed actions. This provides an objective basis for subsequent assessment and incentive mechanisms, and enhances the enthusiasm and reliability of resource collaboration within the park.

[0030] S7. Based on the deviation responsibility weight, the assessment loss ratio is allocated to the corresponding adjustable resource units, a unit-level performance evaluation report is generated, and the initial value of the risk control strategy in the decision-making model is updated and optimized. Furthermore, the total amount of subsidies deducted due to failure to meet the response standards will be allocated to each responsible unit according to the calculated impact weight; For devices that bear a higher deviation weight, their scheduling priority will be reduced in the next round of response tasks; By calling upon the full-process simulation data stored in the digital twin system, the key time periods and dominant factors in which deviations occurred can be analyzed; The deviation correction coefficient in the baseline load forecasting model was revised, and the growth function shape of the risk penalty weight in the rolling optimization was adjusted. It should be noted that by allocating assessment losses to each unit according to responsibility weights, a closed-loop accountability mechanism is formed, which effectively improves the responsiveness and execution accuracy of each equipment entity. At the same time, by combining the results of digital twin simulation backtracking to update the parameters of the baseline prediction model and risk penalty function, the decision-making method achieves continuous self-learning and performance evolution, thereby improving the system's adaptability and overall benefits in multi-cycle demand response tasks.

[0031] This embodiment also provides a computer device applicable to the decision-making method for industrial park demand response considering deviation assessment, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the decision-making method for industrial park demand response considering deviation assessment as proposed in the above embodiment.

[0032] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0033] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the decision-making method for industrial park demand response that considers deviation assessment as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0034] In summary, this invention achieves refined modeling and linear solution of the demand response benefit structure by constructing a segmented subsidy quantification model that considers deviation assessment. Combined with a rolling optimization mechanism and multi-resource collaborative scheduling such as energy storage and interruptible loads, it effectively improves the economy and robustness of the response process. The introduction of digital twin simulation verification and multi-condition disturbance testing enhances the stability and security of the control scheme under abnormal scenarios. The responsibility decomposition mechanism based on marginal contribution attribution achieves fair quantification of deviation responsibility and reasonable allocation of assessment losses. Furthermore, through continuous feedback optimization of prediction accuracy and control strategy, it improves the compliance, autonomous decision-making ability, and overall operational efficiency of industrial parks participating in demand response.

[0035] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A decision-making method for demand response in industrial parks that considers deviation assessment, characterized in that: include: Acquire forecast data on power grid commands, electricity prices, load and renewable energy output of industrial parks during future demand response cycles, as well as operating data of energy storage and interruptible load equipment, and construct a set of adjustable resource units; Based on the set of adjustable resource units, an optimization decision-making model is constructed with the goal of reducing overall operating costs. The optimization decision-making model quantifies the demand response subsidy in segments according to the deviation range between the actual response amount and the target amount, and realizes the linear expression of the segmented revenue structure through logical constraints. During the response execution process, rolling optimization is initiated according to the time step. At each optimization node, the actual operating data that has occurred is called to assess the current cumulative deviation. Combining the energy storage energy change pattern and the evolution trend of system regulation capability, the impact of deviation on the availability of subsequent adjustable resources is predicted, and the deviation impact prediction result is output. Based on the prediction results of the deviation impact, the risk control strategy of the decision model is dynamically adjusted at the current optimization node to generate a corrected control scheme. The modified control scheme is input into a digital twin simulation environment synchronized with the physical system. Multi-condition disturbance tests are conducted using a model that includes the power grid topology and equipment dynamic characteristics to verify the stability and boundary compliance of the scheme under abnormal scenarios. After the response is executed, based on the actual output curve and the planned output curve of the adjustable resource unit set, the marginal contribution attribution method is used to quantify the causal contribution of the adjustable resource unit set to the response deviation, and the deviation responsibility weight of the adjustable resource unit set is output. Based on the deviation responsibility weight, the assessment loss ratio is allocated to the corresponding adjustable resource units, generating a unit-level performance evaluation report, and updating and optimizing the initial value of the risk control strategy in the decision-making model.

2. The decision-making method for industrial park demand response considering deviation assessment as described in claim 1, characterized in that: The specific steps for acquiring forecast data on power grid commands, electricity prices, load, and renewable energy output for the industrial park during future demand response cycles, as well as operational data from energy storage and interruptible load devices, and constructing a set of adjustable resource units are as follows: The peak shaving or valley filling command signals for each time period within the future response time are obtained from the power grid dispatching platform. These command signals are binary variables. When the... The value is 1 when peak-shaving response is required, and 0 otherwise, denoted as . ; The value is 1 when a valley-filling response is required, and 0 otherwise. ; Obtain time-of-use electricity prices for future periods from the electricity market system. ; Based on historical electricity consumption data and weather factors, the baseline power of the industrial park under no-response conditions is calculated. ; The photovoltaic power generation forecast values ​​for different time periods are obtained through the photovoltaic power plant monitoring system. ; Obtain the rated capacity, maximum charge / discharge power, and charging efficiency of the energy storage system. Discharge efficiency Upper and lower limits of state of charge and and initial state of charge ; The adjustable load set, initial power, response power, maximum number of calls per day, and set of allowed response periods are obtained from the interruptible load control system and used to establish the start-stop logic variables, cumulative call count constraints, and time period feasibility constraints for the interruptible load in the optimization model.

3. The decision-making method for industrial park demand response considering deviation assessment as described in claim 2, characterized in that: The aforementioned optimization decision-making model, based on a set of adjustable resource units, aims to reduce overall operating costs. This model quantifies demand response subsidies by segmenting them according to the deviation range between the actual response amount and the target amount, and uses logical constraints to achieve a linear expression of the segmented revenue structure. The specific steps are as follows: The objective function is set to minimize the overall operating cost, and its expression is: ； in, for Electricity price at any time for Total power of the industrial park at any time To reduce peak response subsidies, To fill the valley response subsidy income; Define actual power consumption The compositional relationship is expressed as: ； in, This represents the actual peak reduction response. This represents the actual valley-filling response. This represents the deviation of power from the baseline during non-demand response periods; Introducing constraints to limit the range of the response, the expression is: in, This is the upper limit coefficient for response capability. and Let the variables be 0-1, representing whether peak shaving or valley filling is needed, and satisfying the following conditions: in, It is a very large positive number used to force the response behavior to occur only when the command is issued.

4. The decision-making method for industrial park demand response considering deviation assessment as described in claim 3, characterized in that: During the response execution process, rolling optimization is initiated according to time steps. At each optimization node, actual operating data that has occurred is retrieved to assess the current cumulative deviation. Combining the energy storage energy change pattern and the evolution trend of system regulation capability, the impact of the deviation on the availability of subsequent adjustable resources is predicted, and the deviation impact prediction result is output. The specific steps are as follows: The rolling optimization cycle is set to 15 minutes, at each optimization trigger time. Read from the start of the response until Actual charging and discharging power of energy storage at different times and ; Calculate the current actual state of charge based on the energy conservation principle of energy storage. Its recursive formula is: in, For time intervals; Compare Compared with the original optimization plan The deviation is calculated using the following expression: ； according to Calculate the maximum discharge capacity and maximum charging capacity that energy storage can provide during the remaining time period; By combining the remaining number of interruptible load calls with the available time period, assess whether the system's overall resilience in subsequent time periods meets the remaining response target.

5. The decision-making method for industrial park demand response considering deviation assessment as described in claim 4, characterized in that: The risk control strategy of dynamically adjusting the decision model at the current optimization node based on the deviation impact prediction results, and generating a modified control scheme, includes the following steps: When the prediction results indicate that the subsequent adjustment capacity is sufficient, the original objective function is maintained unchanged; When the prediction results show that the adjustment capability is limited, a bias penalty term is introduced into the objective function, and the adjusted objective function is: in, For the first Response deviation at any moment This is a penalty weighting coefficient that increases over time, used to enhance the suppression of bias in later stages; The optimization model is re-solved to generate a new energy storage charging and discharging plan and an interruptible load scheduling sequence. The revised control scheme is output to update the control instructions for periods that have not been executed.

6. The decision-making method for industrial park demand response considering deviation assessment as described in claim 5, characterized in that: The modified control scheme is input into a digital twin simulation environment synchronized with the physical system. Multi-condition disturbance tests are conducted using a model that includes the power grid topology and equipment dynamic characteristics to verify the stability and boundary compliance of the scheme under abnormal scenarios. The specific steps are as follows: Establish a topology model of the park's power distribution network in the digital twin system, including transformers, feeder impedance, load nodes, and the location of distributed power source access; Configure a dynamic response model for the energy storage system to simulate its charge / discharge switching delay and efficiency degradation characteristics; The modified control scheme is loaded into the simulation model as an input command; Perturbations are applied during the simulation, including scenarios such as sudden drop in photovoltaic output, interruptible load failure, and communication delay. Monitor tie-line power, node voltage, and response trajectory during the simulation process to determine whether the operating boundary constraints are met.

7. The decision-making method for industrial park demand response considering deviation assessment as described in claim 6, characterized in that: After the response is executed, based on the actual output curve and planned output curve of the adjustable resource unit set, the marginal contribution attribution method is used to quantify the causal contribution of the adjustable resource unit set to the response deviation, and the deviation responsibility weight of the adjustable resource unit set is output. The specific steps are as follows: The actual and planned power of energy storage, interruptible loads, and distributed photovoltaic power during each time period within the response cycle; Using all resource combinations as the sample space, calculate the total response of the system under different subset combinations; Based on the Shapley value formula, the marginal response increment of each resource in all possible combination sequences is averaged to obtain its fair contribution to the total response. The overall response deviation is allocated according to the proportion of each resource contribution value to determine the deviation share that each unit should bear.

8. The decision-making method for industrial park demand response considering deviation assessment as described in claim 7, characterized in that: The specific steps for allocating the assessment loss proportion to the corresponding adjustable resource units based on deviation responsibility weights, generating unit-level performance evaluation reports, and updating and optimizing the initial values ​​of risk control strategies in the decision-making model are as follows: The total amount of subsidies deducted due to failure to meet the response standards will be allocated to each responsible unit according to the calculated impact weights. For devices that bear a higher deviation weight, their scheduling priority will be reduced in the next round of response tasks; By calling upon the full-process simulation data stored in the digital twin system, the key time periods and dominant factors in which deviations occurred can be analyzed; The deviation correction coefficient in the baseline load forecasting model is revised, and the growth function shape of the risk penalty weight in the rolling optimization is adjusted.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the decision-making method for industrial park demand response that considers deviation assessment as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the decision-making method for industrial park demand response that takes into account deviation assessment as described in any one of claims 1 to 8.

Images