A multi-dimensional power market optimization control method based on quantified attribution of energy storage

Abstract

The application provides a kind of energy storage participation multi-dimensional power market optimization control method based on quantification attribution.The application carries out initial configuration, constructs energy storage participation multi-dimensional power market clearing model, obtains benchmark scheme and optimal scheme, quantification attribution based on path integral, optimizes control energy storage decision according to attribution result.The application specifically designs the path construction method for mapping marginal benefit using optimal Lagrange multiplier information, proposes the benefit decomposition attribution scheme of riemann approximation based on the path construction method, and generates the power distribution instruction and declaration strategy of energy storage system in the next scheduling period by taking the attribution result as feedback signal.The application can solve the decoupling problem of energy storage value under the coupling of complex physical and market constraints, clearly quantifies the contribution share of energy storage in multi-dimensional power market, optimizes the energy storage scheduling scheme, and provides support for energy storage investment decision and safe and economic operation of new power system.

Description

Technical Field

[0001] This invention belongs to the field of power system dispatch and control, and specifically proposes a method for optimizing control of energy storage participation in multidimensional power market based on quantitative attribution. Background Technology

[0002] With the deepening of the electricity spot market construction and the market-oriented reform of new energy, the functional positioning of energy storage has evolved from the early single function of peak shaving and valley filling to a diversified market player participating in energy market arbitrage and ancillary services such as frequency regulation and reserve. However, with the high proportion of new energy integration, the operational constraints of the power system are becoming increasingly complex. Under the complex coupling of market rules and physical environment, how to achieve accurate quantitative assessment and decoupled control of the multi-dimensional regulation value of energy storage has become a core challenge for resource optimization. Existing technologies usually use black-box optimization models to solve energy storage operation strategies. Although they can provide the overall optimal result, they lack in-depth analysis of the source of the optimization objective, making it difficult to answer the specific contribution share of various regulation services to the overall system benefit, and also difficult to reveal the marginal driving mechanism of different physical and market constraints on the formation of control strategies. This control mode, which lacks interpretability, makes it difficult for the system to generate accurate feedback signals based on the contribution of each constraint. When facing multi-dimensional demand conflicts such as energy and reserve, it cannot provide the optimal weight allocation and dynamic adjustment scheme, limiting the synergistic optimization capability of energy storage resources in ensuring grid security and improving operational efficiency. Summary of the Invention

[0003] The present invention aims to at least partially solve one of the technical problems in the related art.

[0004] Therefore, the first objective of this invention is to propose a quantitative attribution-based method for energy storage participation in multidimensional power market optimization control. Based on multidimensional market optimization clearing models, nearest feasible solution recovery, optimal path construction, path integral calculation, and other technologies, this method uses optimal Lagrange multiplier information to establish a mapping relationship between energy storage decision variables and the reduction of system operating costs. On this basis, the attribution results are used as feedback signals to generate power allocation instructions and application strategies for the energy storage system in the next scheduling cycle.

[0005] The second objective of this invention is to propose a quantitative attribution-based energy storage participation multidimensional electricity market optimization control system.

[0006] To achieve the above objectives, a first aspect of the present invention proposes a method for optimizing control of energy storage participation in a multidimensional electricity market based on quantitative attribution, comprising the following steps: Acquire power system operation data and construct a clearing model for energy storage to participate in the multidimensional electricity market. The clearing model aims to minimize system operating costs and includes energy balance constraints and reserve capacity constraints. The optimal solution for energy storage decision variables is obtained based on the clearing model, and a benchmark solution for energy storage participation in the market is set to determine the evolution path from the benchmark solution to the optimal solution. The marginal benefit mapping relationship along the evolution path is constructed using the optimal Lagrange multiplier information, and the total potential benefit brought by energy storage is decomposed into the contribution share of each constraint to each decision variable through path integral calculation, generating quantitative attribution results. The adjustment preference coefficients for the energy market and the reserve market are calculated based on the quantitative attribution results, and the energy storage charging and discharging power command and market application strategy for the next scheduling cycle are dynamically adjusted based on the adjustment preference coefficients.

[0007] In one embodiment of the present invention, the step of acquiring power system operation data and constructing a clearing model for energy storage participation in a multi-dimensional electricity market includes: Read the system structure parameters of generators, energy storage, AC / DC lines and load nodes, and collect wind, solar and load time series data as boundary conditions; A clearing model is constructed with the objective function of minimizing system operating costs. The objective function is expressed as: the sum of the products of the marginal generation cost of all units and their output in the corresponding time period, plus the product of the increase in reserve deficit and the increase in reserve deficit penalty coefficient, plus the product of the decrease in reserve deficit and the decrease in reserve deficit penalty coefficient. The clearing model is configured with energy market power balance constraints, as well as constraints to increase and decrease reserve demand, resulting in a clearing model that includes energy balance constraints and reserve capacity constraints.

[0008] In one embodiment of the present invention, setting energy market power balance constraints, upward reserve demand constraints, and downward reserve demand constraints in the clearing model includes: For thermal power units, an online capacity continuous variable model is constructed. The output power is limited to between the minimum and maximum technical output by constraints, and the ramp rate is limited to not exceed the maximum ramp rate by constraints. The impact of start-stop operation on output capacity is described by the product relationship between start-stop state variables and maximum output. An operational constraint model is constructed for energy storage devices. Net power is defined by the difference between charging power and discharging power. The energy storage state is updated by adding the net power to the previous energy level and multiplying by the time step. The constraints for adjusting the reserve capacity of energy storage are set as follows: the upward adjustment of reserve capacity shall not exceed the discharging power minus the current net power, and the downward adjustment of reserve capacity shall not exceed the current net power minus the negative minimum charging power, thus forming a complete clearing model that includes energy balance constraints and reserve capacity constraints.

[0009] In one embodiment of the present invention, the step of obtaining the optimal solution for energy storage decision variables based on the clearing model, setting a benchmark solution for energy storage participation in the market, and determining the evolution path from the benchmark solution to the optimal solution includes: Solving the clearing model yields the optimal solution for energy storage decision variables, and the case where energy storage does not participate in the market or only participates in the energy market is set as the benchmark solution; Determine whether the benchmark solution satisfies the market clearing constraint. If the benchmark solution is not feasible, construct the dual problem of the original problem and apply the simplex algorithm to obtain the polar direction. Iterate based on feasible cuts until the nearest feasible solution is obtained. Based on the cost difference between the nearest feasible solution and the optimal solution, the optimal directional path is constructed using the optimal Lagrange multiplier information, and the evolution path from the baseline solution to the optimal solution is determined.

[0010] In one embodiment of the present invention, the step of constructing the optimal directional path using the optimal Lagrange multiplier information includes: The first-order approximate expression of the objective function is obtained by using the optimal Lagrange multipliers. The analytical expression of the optimal direction of the fastest descent is obtained by differentiating the first-order approximate expression. Expand the optimal direction analytical expression element by element, where each term represents the potential benefit of increasing the first type of energy storage decision variable due to the scarcity of the first type of constraint at the current point. A new solution is obtained by advancing a preset step size along the optimal direction. If the new solution is not feasible, the process returns to the step of restoring the most recently feasible solution. Through continuous iteration, the evolution path from the baseline solution to the optimal solution is obtained.

[0011] In one embodiment of the present invention, the step of constructing a marginal benefit mapping relationship along the evolution path using optimal Lagrange multiplier information, and decomposing the total potential benefit brought by energy storage into the contribution share of each constraint to each decision variable through path integral calculation, and generating quantitative attribution results, includes: Integrating the marginal potential benefit term generated by the decision along the evolution path yields a two-dimensional attribution unit; The two-dimensional attribution unit is calculated using the Riemann approximation method of piecewise summation. The corresponding calculation formula is: sum the products of the marginal benefits at all path sampling points and the step size, where the number of sampling points is a preset value. All two-dimensional attribution units are combined to generate a quantitative attribution matrix. The potential benefits of each decision regarding energy storage participation in the market are obtained by summing the columns of the quantitative attribution matrix, forming a quantitative attribution result that includes the contribution share of each constraint to each decision variable.

[0012] In one embodiment of the present invention, the step of calculating the adjustment preference coefficients for the energy market and the reserve market based on the quantitative attribution results, and dynamically adjusting the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficients, includes: The quantitative attribution results are decomposed and classified according to the market dimension of energy storage decision variables, and the attribution values ​​of the standby market and the energy market are extracted. Calculate the adjustment preference coefficient at the current moment, which reflects the system's current demand preference for energy and reserves; When the adjustment preference coefficient indicates that reserve demand is dominant, a capacity shifting strategy is executed to reduce the absolute value of the baseline charging and discharging power to reserve dynamic reserve space. When the adjustment preference coefficient indicates that energy demand is dominant, a full-power operation strategy is executed to prioritize meeting the energy delivery demand and generate the energy storage charging and discharging power command for the next scheduling cycle.

[0013] In one embodiment of the present invention, the step of dynamically adjusting the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficient further includes: Calculate the total attribution value based on the quantitative attribution results, and set the performance judgment threshold; When the total attribution value is greater than the performance determination threshold, it is determined to be a high-performance period. The optimized control system relaxes the declaration boundary to allow for deeper charge and discharge depths, ensuring that this period is prioritized by the market clearing model. When the total attribution value is less than or equal to the performance determination threshold, it is determined to be a period of low performance. The optimized control system shrinks the declaration boundary to guide the scheduling system to reduce the call to energy storage during this period, thereby preserving the lifespan of energy storage and generating a market declaration strategy that includes instructions to relax or shrink the declaration boundary.

[0014] To achieve the above objectives, a second aspect of the present invention proposes a quantitative attribution-based multidimensional power market optimization control system for energy storage participation, comprising: The clearing model construction module is used to acquire power system operation data and construct a clearing model for energy storage to participate in the multidimensional power market. The clearing model aims to minimize system operating costs and includes energy balance constraints and reserve capacity constraints. The path determination module is used to solve for the optimal solution of energy storage decision variables based on the clearing model, set the benchmark solution for energy storage to participate in the market, and determine the evolution path from the benchmark solution to the optimal solution. The quantitative attribution result generation module is used to construct the marginal benefit mapping relationship along the evolution path using the optimal Lagrange multiplier information, and to decompose the total potential benefit brought by energy storage into the contribution share of each constraint to each decision variable through path integral calculation, thereby generating quantitative attribution results. The dynamic adjustment module is used to calculate the adjustment preference coefficients of the energy market and the reserve market based on the quantitative attribution results, and dynamically adjust the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficients.

[0015] The methods and systems of this invention can accurately decompose the total potential benefits of energy storage into various market dimensions, key constraints, and decision variables, significantly enhancing the interpretability of control strategies, clearly quantifying the contribution share of energy storage in multi-dimensional scenarios, and optimizing scheduling schemes accordingly. This provides scientific quantitative and control support for energy storage investment decisions, revenue settlement, and safe and economical operation of new power systems.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a method for optimizing and controlling multidimensional electricity market participation based on quantitative attribution, provided as an embodiment of the present invention; Figure 2 This is a structural diagram of a multi-dimensional power market optimization control system based on quantitative attribution, provided as an embodiment of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] The following description, with reference to the accompanying drawings, describes a method and system for optimizing and controlling multidimensional electricity market participation based on quantitative attribution, according to an embodiment of the present invention.

[0021] Figure 1 This is a flowchart illustrating a method for optimizing the control of multidimensional electricity market participation based on quantitative attribution, according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps: S1. Obtain power system operation data and construct a clearing model for energy storage to participate in the multidimensional power market. The clearing model aims to minimize system operating costs and includes energy balance constraints and reserve capacity constraints. S2, Based on the clearing model, the optimal solution for energy storage decision variables is obtained, and a benchmark solution for energy storage participation in the market is set, and the evolution path from the benchmark solution to the optimal solution is determined; S3. Construct the marginal benefit mapping relationship along the evolution path using the optimal Lagrange multiplier information, and decompose the total potential benefit brought by energy storage into the contribution share of each constraint to each decision variable through path integral calculation, and generate quantitative attribution results. S4. Calculate the adjustment preference coefficients for the energy market and the reserve market based on the quantitative attribution results, and dynamically adjust the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficients.

[0022] Specifically, before implementing this invention, S0 is required to perform initial configuration.

[0023] Read system structure and equipment parameters, including information on generators, energy storage, AC / DC lines, and load nodes. Collect and input relevant parameters such as generator parameters, energy storage parameters, and line capacity. Read time-series data of wind, solar, and load as boundary conditions.

[0024] Step S1: Construct a multi-dimensional electricity market clearing model for energy storage participation.

[0025] Taking energy storage participation in the spot energy market and the standby capacity market as an example, the objective function is to minimize system costs and add penalties for standby deficits.

[0026]

[0027] in It is the marginal cost of generating electricity from the unit. It is the output of unit g during the time period. Adjust the reserve deficit upwards / downwards. Adjust the penalty coefficient for reserve shortfalls upwards or downwards.

[0028] The energy market requires power balance:

[0029] in The charging and discharging power of energy storage, The load of the node.

[0030] For conventional thermal power units, the corresponding discrete variables are linearized, that is, the capacity changes caused by the unit's start-up and shutdown actions are represented by the unit's online capacity. This is reflected in continuous variable modeling.

[0031]

[0032] Corresponding to the traditional unit combination, Equation (3) sets upper and lower limits for the output power of the thermal power unit. Equation (4) specifies the ramp rate limit of the unit, requiring that the power adjustment range of the thermal power unit shall not exceed its maximum allowable upward or downward adjustment rate. Equation (5) is used to describe the start-up and shutdown operations of the thermal power unit at different times. Its changes are modeled through the online capacity of the unit, accurately reflecting the impact of start-up and shutdown on the unit's output capacity. Equations (6) and (7) are used to constrain the minimum continuous start-up time and minimum continuous shutdown time of the unit, respectively, to ensure that each thermal power unit can meet the minimum duration requirement during operation and shutdown.

[0033] For renewable energy units, curtailment of wind and solar power is permitted:

[0034] in, For wind power and photovoltaic units, It reflects the per-unit value of the maximum output of renewable energy.

[0035] The operating constraints of energy storage devices are as follows:

[0036] in, The efficiency of charging and discharging energy storage devices. The maximum power discharge duration, time step This indicates the duration for which energy storage is charged or discharged at a constant power.

[0037] In the standby market, the upward adjustment of standby demand satisfies:

[0038] in, To increase reserve requirements.

[0039] In the standby market, reducing standby demand satisfies:

[0040] in, To reduce reserve requirements.

[0041] The standby time for thermal power units is as follows:

[0042] The up and down reserves of energy storage are as follows:

[0043] For ease of explanation, the above clearing model is simplified to the following vector matrix form. Where Represents decision variables related to energy storage. This represents the remaining decision variables. Correspondingly, the parameter matrix can be divided into two parts. This represents the optimal Lagrange multiplier corresponding to the constraint condition.

[0044]

[0045] Step S2: Obtain the baseline solution and the optimal solution.

[0046] Suppose that the optimal solution to problem (21) corresponds to the optimal energy storage decision scheme. :

[0047] Given a baseline scheme (Energy storage does not participate in the market, energy storage only participates in the energy market, etc.), conduct a quantitative attribution analysis of energy storage participation in the multi-dimensional electricity market relative to this benchmark scheme, that is, construct and obtain the attribution matrix:

[0048] Where m is the number of constraints and n is the number of decision variables. This represents the share of improvement in the market clearing outcome relative to the benchmark objective function contributed by energy storage decision variable j under constraint i. This potential decrease in performance is termed the "potential benefit" of energy storage participation in the market. By summing the results, we can obtain the potential benefits of various decisions regarding energy storage participation in the market.

[0049]

[0050] In the formula It is an m-dimensional vector whose elements are all 1.

[0051] It is important to emphasize that the existence of a benchmark scheme is necessary: ​​the potential benefits of energy storage participating in the multidimensional electricity market are essentially relative improvements. Only by comparing with the benchmark scheme can there be a quantifiable object for cost reduction / profit increase.

[0052] Step S3: Quantitative attribution based on path integral.

[0053] 1) Sub-step S31: Recover the most recently feasible solution Without loss of generality, we assume the baseline scheme Infeasible: This means that under the benchmark energy storage decision scheme, the market clearing constraint cannot be met, thus making the problem infeasible. In this case, a portion of the variable energy storage decision process evolving from the benchmark scheme to the optimal scheme is necessary to first restore market feasibility. Let this portion result in the following scheme: This is referred to as the closest feasible solution.

[0054] To obtain directional information from infeasible points to feasible regions, we can construct the dual problem of the primal problem:

[0055] Solving the above problem using the simplex algorithm can yield the polar direction. Adjusting the right-hand side in the opposite direction can make the problem more feasible. Add the following feasible cut:

[0056] correspond The coefficient is the feasible direction:

[0057] To obtain the most recent feasible solution Solve the following problem:

[0058] If the solution obtained by solving the above equation is still infeasible, then generate feasible cuts again and add them, iterating until a feasible solution is found.

[0059] 2) Sub-step S32: Constructing the optimal direction path After obtaining feasible solutions, each solution within the feasible region corresponds to an objective function value. The total potential benefit of the baseline solution relative to the optimal solution is:

[0060] In order to Attribution, introduction At point gradient . It is an n-dimensional vector. This represents the marginal impact of a unit increase in decision dimension j on the total potential benefit. This is achieved using the optimal Lagrange multiplier. available First-order approximation expression:

[0061] right Differentiating the equation yields the analytical expression for the direction of the fastest descent, i.e., the optimal direction:

[0062] Will Expanding by elements, we can obtain the marginal benefit derived from the shadow prices of each constraint. pass Mapped to the energy storage decision-making dimension:

[0063] in This can be explained as follows: due to the scarcity of the i-th type of constraint at the current point. This leads to the potential benefits of increasing the j-th type of energy storage decision variables.

[0064] By moving a certain number of steps in the optimal direction, a new solution with a lower total cost is obtained:

[0065] If the new solution is not feasible, return to sub-step S31 to restore the most recently feasible solution, and continue iterating to obtain the solution from the previous one. arrive Evolutionary path .

[0066] 3) Sub-step S33: Path integral attribution Along the path Integrating the marginal potential benefit term generated by decision j for constraint i, we obtain the two-dimensional attribution unit. :

[0067] In practice, path integrals cannot be calculated precisely, so a piecewise summation Riemann approximation method is used:

[0068] Where K is the number of path sampling points, The smaller the value of K, the larger the approximation, and the more accurate the approximation; because The essence is the allocation of total potential benefits. Ideally, it should satisfy closure:

[0069] The above formula can be used to estimate the approximate error of the product summation result relative to the true value. In practice, the step size can be adjusted according to the magnitude of the error; the vector magnitude between the nearest feasible solution and the optimal solution can be used as a reference for setting the step size. If the difference between the current solution and the optimal solution is less than the step size... If so, the solution for the next iteration is set as the optimal solution, and the potential benefit value of the attribution is updated for the last time.

[0070] Step S4: Optimize control energy storage decisions based on attribution results.

[0071] After completing the path integral and quantified attribution, the contribution of each energy storage decision to the total potential benefit is obtained by summing the columns of the attribution matrix. Furthermore, the attribution value of energy storage in each market is decomposed and classified according to the market dimension of the energy storage decision variables, and the various contributions of energy storage in the multidimensional electricity market are quantified.

[0072] Based on this, and according to the decoupling attribution results of energy and reserve dimensions, the adjustment preference coefficient at the current moment is calculated. .

[0073]

[0074] in, As the attribution value for the standby market, This is the energy market attribution value. This reflects the system's current energy and reserve demand preferences. The control system adjusts the active power command for the next scheduling cycle based on the preference coefficient: For scenarios dominated by reserve demand, the control system implements a capacity shifting strategy, proactively reducing the absolute value of the baseline charging and discharging power, thereby reserving more dynamic reserve space to improve the system's instantaneous response capability to frequency regulation or reserve commands. For scenarios dominated by energy demand, the physical adjustment value of energy time shifting is greatest during this period. The control system implements a full-power operation strategy, prioritizing the fulfillment of power delivery needs, ensuring that the energy storage device accurately tracks the active power setpoint, and maximizing the peak shaving and valley filling effect on the grid load.

[0075] When considering reporting strategies, the optimization control system uses the total attribution value from the attribution results. The reporting parameters for the next scheduling cycle are dynamically adjusted. A threshold is set. ,for During peak energy consumption periods, energy storage output regulation is crucial for achieving system-wide optimization. Optimizing the control system by relaxing application boundaries and allowing for deeper charge / discharge depths ensures that these periods are prioritized by the market clearing model. During periods of low energy efficiency, energy storage regulation contributes little to the system. Optimizing the control system by narrowing the reporting boundary guides the scheduling system to reduce the use of energy storage during this period, thereby preserving the lifespan of the energy storage.

[0076] The energy storage participation multidimensional electricity market optimization control method based on quantitative attribution according to embodiments of the present invention can quantitatively evaluate the value contribution of energy storage in various multidimensional markets. By constructing clearing models for energy storage participation in the spot energy market and reserve capacity market, and using path integral technology to accurately decompose the total potential benefits brought by energy storage relative to the benchmark scheme, the charging and discharging power of energy storage and reserve capacity reservation are dynamically adjusted accordingly. This achieves optimal market control of energy storage resources while ensuring the safe operation of the power grid. The method involves initial configuration, construction of a clearing model for energy storage participation in the multidimensional electricity market, acquisition of benchmark and optimal schemes, quantitative attribution based on path integral, and optimization control of energy storage decisions based on the attribution results. Specifically, the present invention designs a path construction method that maps marginal benefits using optimal Lagrange multiplier information. Based on this, it proposes a benefit decomposition attribution scheme based on Riemann approximation, and uses the attribution results as feedback signals to generate power allocation instructions and application strategies for the energy storage system in the next scheduling cycle. This invention can solve the problem of decoupling the value of energy storage under the coupling of complex physics and market constraints, clearly quantify the contribution share of energy storage in the multidimensional electricity market, optimize the control of energy storage dispatch schemes, and provide support for energy storage investment decisions and safe and economical operation of new power systems.

[0077] To implement the method of the embodiments of the present invention, the present invention also proposes a quantitative attribution-based energy storage participation multidimensional electricity market optimization control system 10, such as... Figure 2 As shown, it includes: The clearing model construction module 100 is used to acquire power system operation data and construct a clearing model for energy storage to participate in the multidimensional power market. The clearing model aims to minimize system operating costs and includes energy balance constraints and reserve capacity constraints. The path determination module 200 is used to solve for the optimal solution of energy storage decision variables based on the clearing model, set the benchmark solution for energy storage to participate in the market, and determine the evolution path from the benchmark solution to the optimal solution. The quantitative attribution result generation module 300 is used to construct the marginal benefit mapping relationship along the evolution path using the optimal Lagrange multiplier information, and to decompose the total potential benefit brought by energy storage into the contribution share of each constraint to each decision variable through path integral calculation, thereby generating quantitative attribution results. The dynamic adjustment module 400 is used to calculate the adjustment preference coefficients of the energy market and the reserve market based on the quantitative attribution results, and dynamically adjust the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficients.

[0078] The energy storage participation multidimensional electricity market optimization control system based on quantitative attribution according to embodiments of the present invention can solve the decoupling problem of energy storage value under the coupling of complex physics and market constraints, clearly quantify the contribution share of energy storage in the multidimensional electricity market, optimize the control of energy storage dispatch schemes, and provide support for energy storage investment decisions and safe and economical operation of new power systems.

[0079] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0080] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0081] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for energy storage participation in multidimensional electricity market optimization control based on quantitative attribution, characterized in that, include: Acquire power system operation data and construct a clearing model for energy storage to participate in the multidimensional electricity market. The clearing model aims to minimize system operating costs and includes energy balance constraints and reserve capacity constraints. The optimal solution for energy storage decision variables is obtained based on the clearing model, and a benchmark solution for energy storage participation in the market is set to determine the evolution path from the benchmark solution to the optimal solution. The marginal benefit mapping relationship along the evolution path is constructed using the optimal Lagrange multiplier information, and the total potential benefit brought by energy storage is decomposed into the contribution share of each constraint to each decision variable through path integral calculation, generating quantitative attribution results. The adjustment preference coefficients for the energy market and the reserve market are calculated based on the quantitative attribution results, and the energy storage charging and discharging power command and market application strategy for the next scheduling cycle are dynamically adjusted based on the adjustment preference coefficients.

2. The method as described in claim 1, characterized in that, The acquisition of power system operation data and the construction of a clearing model for energy storage participation in the multidimensional electricity market include: Read the system structure parameters of generators, energy storage, AC / DC lines and load nodes, and collect wind, solar and load time series data as boundary conditions; A clearing model is constructed with the objective function of minimizing system operating costs. The objective function is expressed as: the sum of the products of the marginal generation cost of all units and their output in the corresponding time period, plus the product of the increase in reserve deficit and the increase in reserve deficit penalty coefficient, plus the product of the decrease in reserve deficit and the decrease in reserve deficit penalty coefficient. The clearing model is configured with energy market power balance constraints, as well as constraints to increase and decrease reserve demand, resulting in a clearing model that includes energy balance constraints and reserve capacity constraints.

3. The method as described in claim 2, characterized in that, The setting of energy market power balance constraints, upward adjustment constraints on reserve demand, and downward adjustment constraints on reserve demand in the clearing model includes: For thermal power units, an online capacity continuous variable model is constructed. The output power is limited to between the minimum and maximum technical output by constraints, and the ramp rate is limited to not exceed the maximum ramp rate by constraints. The impact of start-stop operation on output capacity is described by the product relationship between start-stop state variables and maximum output. An operational constraint model is constructed for energy storage devices. Net power is defined by the difference between charging power and discharging power. The energy storage state is updated by adding the net power to the previous energy level and multiplying by the time step. The constraints for adjusting the reserve capacity of energy storage are set as follows: the upward adjustment of reserve capacity shall not exceed the discharging power minus the current net power, and the downward adjustment of reserve capacity shall not exceed the current net power minus the negative minimum charging power, thus forming a complete clearing model that includes energy balance constraints and reserve capacity constraints.

4. The method as described in claim 1, characterized in that, The process of obtaining the optimal solution for energy storage decision variables based on the clearing model, setting a benchmark solution for energy storage participation in the market, and determining the evolution path from the benchmark solution to the optimal solution includes: Solving the clearing model yields the optimal solution for energy storage decision variables, and the case where energy storage does not participate in the market or only participates in the energy market is set as the benchmark solution; Determine whether the benchmark solution satisfies the market clearing constraint. If the benchmark solution is not feasible, construct the dual problem of the original problem and apply the simplex algorithm to obtain the polar direction. Iterate based on feasible cuts until the nearest feasible solution is obtained. Based on the cost difference between the nearest feasible solution and the optimal solution, the optimal directional path is constructed using the optimal Lagrange multiplier information, and the evolution path from the baseline solution to the optimal solution is determined.

5. The method as described in claim 4, characterized in that, The construction of the optimal directional path using optimal Lagrange multiplier information includes: The first-order approximate expression of the objective function is obtained by using the optimal Lagrange multipliers. The analytical expression of the optimal direction of the fastest descent is obtained by differentiating the first-order approximate expression. Expand the optimal direction analytical expression element by element, where each term represents the potential benefit of increasing the first type of energy storage decision variable due to the scarcity of the first type of constraint at the current point. A new solution is obtained by advancing a preset step size along the optimal direction. If the new solution is not feasible, the process returns to the step of restoring the most recently feasible solution. Through continuous iteration, the evolution path from the baseline solution to the optimal solution is obtained.

6. The method as described in claim 1, characterized in that, The method involves constructing a marginal benefit mapping relationship along the evolution path using optimal Lagrange multiplier information, and decomposing the total potential benefit brought by energy storage into the contribution share of each constraint to each decision variable through path integral calculation, generating quantitative attribution results, including: Integrating the marginal potential benefit term generated by the decision along the evolution path yields a two-dimensional attribution unit; The two-dimensional attribution unit is calculated using the Riemann approximation method of piecewise summation. The corresponding calculation formula is: sum the products of the marginal benefits at all path sampling points and the step size, where the number of sampling points is a preset value. All two-dimensional attribution units are combined to generate a quantitative attribution matrix. The potential benefits of each decision regarding energy storage participation in the market are obtained by summing the columns of the quantitative attribution matrix, forming a quantitative attribution result that includes the contribution share of each constraint to each decision variable.

7. The method as described in claim 1, characterized in that, The step of calculating the adjustment preference coefficients for the energy market and the reserve market based on the quantitative attribution results, and dynamically adjusting the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficients, includes: The quantitative attribution results are decomposed and classified according to the market dimension of energy storage decision variables, and the attribution values ​​of the standby market and the energy market are extracted. Calculate the adjustment preference coefficient at the current moment, which reflects the system's current demand preference for energy and reserves; When the adjustment preference coefficient indicates that reserve demand is dominant, a capacity shifting strategy is executed to reduce the absolute value of the baseline charging and discharging power to reserve dynamic reserve space. When the adjustment preference coefficient indicates that energy demand is dominant, a full-power operation strategy is executed to prioritize meeting the energy delivery demand and generate the energy storage charging and discharging power command for the next scheduling cycle.

8. The method as described in claim 7, characterized in that, The method of dynamically adjusting the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficient also includes: Calculate the total attribution value based on the quantitative attribution results, and set the performance judgment threshold; When the total attribution value is greater than the performance determination threshold, it is determined to be a high-performance period. The optimized control system relaxes the declaration boundary to allow for deeper charge and discharge depths, ensuring that this period is prioritized by the market clearing model. When the total attribution value is less than or equal to the performance determination threshold, it is determined to be a period of low performance. The optimized control system shrinks the declaration boundary to guide the scheduling system to reduce the call to energy storage during this period, thereby preserving the lifespan of energy storage and generating a market declaration strategy that includes instructions to relax or shrink the declaration boundary.

9. A multi-dimensional electricity market optimization control system based on quantitative attribution for energy storage participation, characterized in that, include: The clearing model construction module is used to acquire power system operation data and construct a clearing model for energy storage to participate in the multidimensional electricity market. The clearing model aims to minimize system operating costs and includes energy balance constraints and reserve capacity constraints. The path determination module is used to solve for the optimal solution of energy storage decision variables based on the clearing model, set the benchmark solution for energy storage to participate in the market, and determine the evolution path from the benchmark solution to the optimal solution. The quantitative attribution result generation module is used to construct the marginal benefit mapping relationship along the evolution path using the optimal Lagrange multiplier information, and to decompose the total potential benefit brought by energy storage into the contribution share of each constraint to each decision variable through path integral calculation, thereby generating quantitative attribution results. The dynamic adjustment module is used to calculate the adjustment preference coefficients of the energy market and the reserve market based on the quantitative attribution results, and dynamically adjust the energy storage charging and discharging power command and market application strategy for the next scheduling cycle based on the adjustment preference coefficients.

Images