Energy storage electric power spot market simulation clearing method based on hierarchical model control prediction

Abstract

The invention relates to an energy storage electric power spot market simulation clearing method based on hierarchical model control prediction, and the method comprises the following steps: S1, building an electric power spot market global simplified clearing model containing energy storage, and taking the model as an upper prediction control model; s2, establishing an electric power spot market detailed clearing model containing energy storage as a lower execution model; s3, dividing long-time simulation into M solving periods; in each solving period, global simplification clearing optimization is firstly utilized to obtain an energy storage final state SOC constraint value of each remaining period; on the basis of the energy storage final state SOC constraint value of the current period, market simulation clearing of the current period is completed by using a detailed clearing model; and finally, updating the system state according to the current clearing result, and rolling to advance to the next cycle until clearing of all cycles is completed. According to the method, dynamic global optimization of the energy storage SOC constraint is realized through a hierarchical model prediction control mechanism, and the calculation efficiency and the market simulation clearing effect are effectively balanced.

Description

Technical Field

[0001] This invention relates to the field of electrical engineering, and more specifically, to a method for simulating clearing in the energy storage electricity spot market based on hierarchical model control prediction. Background Technology

[0002] Energy storage systems, with their flexible adjustment capabilities, can play a vital role in the market, such as peak shaving, valley filling, congestion mitigation, and price price fluctuation suppression. Especially in the electricity spot market, energy storage can quickly respond to sudden changes in renewable energy output and load surges, effectively smoothing price spikes and enhancing system stability, playing a crucial role in improving the overall market efficiency and economics. Simulated clearing in the electricity spot market is an important method for simulating electricity market operations and reflecting the market behavior and value of various generating units. It involves a large number of unit combinations and constraints for various units, while also addressing the uncertainties of load and renewable energy output.

[0003] However, energy storage still faces significant challenges in participating in long-term simulated clearing of the electricity spot market. While decomposing the long-term global optimization problem into multiple solution cycles with rolling clearing can reduce computational complexity, it is difficult to reasonably determine the state of charge (SOC) constraint value of energy storage at the end of each solution cycle. Without constraints, energy storage dispatch strategies may lack a global perspective and exhibit short-sighted behavior—such as over-discharging in the current cycle, resulting in insufficient regulation capacity at subsequent critical moments, inability to cope with price peaks or system disturbances, and reduced overall economic efficiency and operational reliability. If each cycle uses equal initial and final SOC or relies on dispatchers' experience to set SOC boundary conditions, there is a lack of systematic optimization basis. This not only leads to strong subjectivity and limited optimization effects but also makes it difficult to adapt to the constantly changing grid conditions and market environment in actual operation, thus restricting the efficient application of energy storage in the spot market. Summary of the Invention

[0004] The technical problem to be solved by this invention is to provide a simulated clearing method for the energy storage electricity spot market based on hierarchical model control prediction. This method effectively solves the final state of charge (SOC) constraint problem of energy storage in long-term simulated clearing of the spot market, and significantly improves the regulatory value of energy storage in the spot market and the reliability of system operation.

[0005] The technical solution adopted by this invention to solve its technical problem is: to construct a simulated clearing method for the energy storage electricity spot market based on hierarchical model control and prediction, including the following steps: S1. Establish a simplified global clearing model for the electricity spot market that includes energy storage, as an upper-level predictive control model; S2. Establish a detailed clearing model for the electricity spot market that includes energy storage, as the underlying execution model; S3. Divide the long-term simulation into M solution cycles. In each solution cycle, first use global simplified clearing optimization to obtain the remaining energy storage final state SOC constraint values ​​for each cycle. Then, based on the energy storage final state SOC constraint values ​​for this cycle, use the detailed clearing model to complete the market simulation clearing for this cycle. Finally, update the system state according to the current clearing results and roll forward to the next cycle until all cycles are cleared.

[0006] According to the above scheme, in step S1, the key constraints retained in the global simplified clearing model of the electricity spot market include system power balance constraints, key section transmission constraints, thermal power unit operation constraints, energy storage operation constraints, and renewable energy output constraints.

[0007] According to the above scheme, the global simplified clearing model is simplified by using an aggregated time scale. The simulation period is set to 1 hour or several hours as needed, and the data for each period is the average of the corresponding spot market 15-minute interval data.

[0008] According to the above scheme, the mathematical model of the global simplified clearing model for the electricity spot market is as follows: The objective function is to minimize the total operating cost of the system.

[0009] Where N is the number of generating units, and T0 is the number of time periods considered. Let be the operating cost of unit i in time period t, which is a piecewise linear function relating the unit's declared output ranges and corresponding energy prices. This represents the net output of unit i during time period t; System power balance constraints:

[0010] in The total system load during time period t; Energy storage operation constraints include upper and lower limits of energy storage charge / discharge power, dynamic evolution constraints of SOC, upper and lower limits of SOC, and final-state SOC constraints of energy storage.

[0011]

[0012]

[0013]

[0014] in, , These represent the discharge power and charging power of the energy storage unit during time period t, respectively. , These are the maximum charging and discharging power and the maximum stored capacity of the energy storage unit, respectively. The energy storage unit is in its state of charge. This represents the simulation time step size; , These represent the minimum and maximum states of charge, respectively. , These are discharge and charge efficiency, respectively. Self-discharge rate; The final-state SOC constraint value for energy storage in the global simulation; Conventional constraints in the spot market, such as key section power flow constraints, thermal power unit operation constraints, and renewable energy output constraints:

[0015]

[0016]

[0017] in, The planned power of critical section j in time period t. Let i be the active power output of thermal power unit i during time period t. For the active power output of renewable energy unit i in time period t, The upper limit of the power flow at the critical section j. , These are the lower and upper limits of the output of thermal power unit i, respectively. Let be the ramp rate of thermal power unit i. This represents the predicted output value of renewable energy unit i during time period t.

[0018] According to the above scheme, in step S2, the detailed clearing model of the electricity spot market with energy storage includes the safety-constrained unit combination model (SCUC) and the safety-constrained economic dispatch model (SCED) of the spot market. First, the start-up and shutdown status and output boundary of conventional units in each period are optimized and determined through the safety-constrained unit combination model (SCUC) of the spot market. Then, under the premise of fixed start-up and shutdown status, the output allocation of each unit is refined and optimized through the safety-constrained economic dispatch model (SCED) to ensure that the network security constraints and economic efficiency requirements are met. The nodal pricing mechanism is used to calculate the clearing price of each node.

[0019] According to the above scheme, the constraints on energy storage operation include upper and lower limits of energy storage charging / discharging power, dynamic evolution constraints of SOC, upper and lower limits of SOC, and final state SOC constraints of energy storage:

[0020]

[0021]

[0022]

[0023] in, , These represent the discharge power and charging power of the energy storage unit during time period t, respectively. , These are the maximum charging and discharging power and the maximum stored capacity of the energy storage unit, respectively. The energy storage unit is in its state of charge. To solve for the time step size; , These represent the minimum and maximum states of charge, respectively. , These are discharge and charge efficiency, respectively. Self-discharge rate; The final state of energy storage in the clearing result is the SOC. This represents the final-state SOC constraint value of the energy storage obtained by clearing the upper-level model.

[0024] According to the above scheme, in step S3, the hierarchical rolling clearing process achieves long-term simulated clearing through a "prediction-execution-update" rolling mechanism. The specific steps are as follows: S301: Divide the long simulation into M solution cycles. The cycle length is determined by taking into account the solution speed and the operation cycle characteristics of energy storage. This forms a three-level time structure of "total simulation time, M solution cycles, and each cycle containing multiple 15-minute time intervals". S302: Upper-level prediction optimization. The solution period of the upper-level global simplified clearing model changes dynamically. The first solution period covers the entire simulation duration. The initial value is the global initial state of the system. After each solution cycle of spot market clearing is completed, the starting point of the global simplified clearing model rolls to the initial time of the current solution cycle, while the ending point remains unchanged. The initial state is the final state of the clearing result of the previous solution cycle. S303: Detailed clearing of the spot market in the next solution cycle. The initial value for the first solution cycle is the global initial state of the system, and the initial values ​​for subsequent solution cycles are the final state of the clearing result of the previous solution cycle, to obtain... The final state SOC constraint value of the energy storage as the solution period is cleared using a detailed clearing model of the spot market to obtain detailed clearing results for each 15-minute period. S304: Update the system state with the current cycle's final state, return to step 302, and repeat the upper-level prediction and lower-level detailed clearing until all M cycles have been cleared.

[0025] According to the above scheme, in step S302, the upper-level optimization problem is solved using a global simplified clearing model to obtain the optimized SOC values ​​of the energy storage final state for each cycle, in the following order: k represents the number of remaining solution cycles in the current global simplified clearing model, which decreases gradually as the model rolls on. Only the SOC optimization value at the final state of the energy storage in the current solution cycle is extracted as the constraint value. Used for execution at the lower level.

[0026] According to the above scheme, the final state of energy storage (SOC) constraint value The following settings are used: The first solution period covers the entire simulation duration, from the initial moment to the end of the Mth solution cycle. The initial value is the global initial state of the system. For each subsequent solution cycle, after the spot market clearing is completed, the starting point of the global simplified clearing model rolls back to the initial moment of the current solution cycle, while the ending point remains unchanged. The initial state is the final state of the clearing result from the previous solution cycle. The global simplified clearing model is used to complete the solution clearing, and the clearing result includes the optimized SOC values ​​for each time period. The optimized SOC value at the final state of the current solution cycle is extracted as the constraint value. Used for execution at the lower level.

[0027] According to the above scheme, in S303, the detailed clearing results include the output of thermal power units, the output of renewable energy, the charging / discharging power of energy storage, the SOC of energy storage, and the nodal price for each time period.

[0028] The energy storage electricity spot market simulation clearing method based on hierarchical model control prediction of the present invention has the following beneficial effects: This invention achieves a balance between "global perspective" and "local refinement" through a hierarchical model predictive control mechanism: the upper-level global simplified clearing model quickly predicts future SOC evolution trends, avoiding overcharging and discharging in the current cycle due to "shortsightedness"; the lower-level detailed clearing model ensures the accuracy and optimization performance of single-cycle clearing; it eliminates the need to rely on manual experience to set SOC constraints, dynamically adapts to market changes, effectively solves the energy storage final-state SOC constraint problem in long-term spot market simulated clearing, and significantly improves the regulatory value of energy storage in the spot market and the reliability of system operation; it realizes dynamic global optimization of energy storage SOC constraints, effectively balancing computational efficiency and market simulated clearing effects. Attached Figure Description

[0029] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart of the simulated clearing method for the energy storage electricity spot market based on hierarchical model control prediction, as described in this invention. Figure 2 This is a schematic diagram of the rolling clearing principle of the hierarchical model control prediction of the present invention; Figure 3This is the total system load curve of the present invention; Figure 4 This is a comparison chart of system node electricity prices between the conventional periodic clearing method and the hierarchical model predictive control method of this invention; Figure 5 This is a comparison chart of typical time periods for system node electricity prices between the conventional periodic clearing method and the hierarchical model predictive control method of this invention; Figure 6 This is a comparison chart of the energy storage SOC between the conventional periodic clearing method and the hierarchical model predictive control method of this invention. Detailed Implementation

[0030] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0031] like Figure 1-6 As shown, the energy storage electricity spot market simulation clearing method based on hierarchical model control prediction of the present invention includes the following steps: S1. Establish a simplified global clearing model for the electricity spot market that includes energy storage, as the upper-level predictive control model.

[0032] The simplified global clearing model for the electricity spot market retains key constraints including system power balance constraints, key section transmission constraints, thermal power unit operation constraints, energy storage operation constraints, and renewable energy output constraints. The simplified global clearing model for the electricity spot market uses an aggregated time scale for simplification, setting the simulation period to one hour or several hours as needed, and using the average of the corresponding 15-minute interval data from the spot market for each period.

[0033] The mathematical model is as follows: The objective function is to minimize the total operating cost of the system.

[0034] Where N represents the number of generating units, including thermal power units, energy storage units, and renewable energy units, and T0 represents the number of time periods considered. Let be the operating cost of unit i in time period t, which is a piecewise linear function relating the unit's declared output ranges and corresponding energy prices. This represents the net output of unit i during time period t, which is negative when energy storage is charging and positive when discharging.

[0035] System power balance constraints:

[0036] in Let t be the total system load during time period t.

[0037] Energy storage operation constraints include upper and lower limits of energy storage charge / discharge power, dynamic evolution constraints of SOC, upper and lower limits of SOC, and final-state SOC constraints of energy storage.

[0038]

[0039]

[0040]

[0041] in, , These represent the discharge power and charging power of the energy storage unit during time period t, respectively. , These are the maximum charging and discharging power and the maximum stored capacity of the energy storage unit, respectively. The energy storage unit is in its state of charge. This represents the simulation time step size; , These represent the minimum and maximum states of charge, respectively. , These are discharge and charge efficiency, respectively. Self-discharge rate; This represents the final-state SOC constraint value for energy storage in the global simulation.

[0042] Conventional constraints in the spot market, such as key section power flow constraints, thermal power unit operation constraints, and renewable energy output constraints:

[0043]

[0044]

[0045] in, The planned power of critical section j in time period t. Let i be the active power output of thermal power unit i during time period t. For the active power output of renewable energy unit i in time period t, The upper limit of the power flow at the critical section j. , These are the lower and upper limits of the output of thermal power unit i, respectively. Let be the ramp rate of thermal power unit i. This represents the predicted output value of renewable energy unit i during time period t.

[0046] S2. Establish a detailed clearing model for the electricity spot market that includes energy storage, as the lower-level execution model, including a safety-constrained unit combination model and a safety-constrained economic dispatch model.

[0047] The detailed clearing model for the electricity spot market with energy storage includes a Security Constrained Unit Combination Model (SCUC) and a Security Constrained Economic Dispatch Model (SCED) for the spot market. First, the start-up and shutdown status and output boundaries of conventional units in each time period within the cycle are determined through SCUC optimization. Then, under the premise of fixed start-up and shutdown status, SCED refines and optimizes the output allocation of each unit with energy storage, ensuring compliance with network security constraints and economic efficiency requirements. The clearing price for each node is calculated using a nodal pricing mechanism. The model uses a 15-minute base time period and fully incorporates the constraints of the spot market, including: system load balance constraints, system positive / negative reserve capacity constraints, system spinning reserve constraints, unit output upper and lower limit constraints, unit ramping constraints, unit minimum continuous start-up and shutdown time constraints, line power flow constraints, cross-sectional power flow constraints, and renewable energy output constraints. The operating constraints for energy storage units are as follows: Energy storage operation constraints include upper and lower limits of energy storage charge / discharge power, dynamic evolution constraints of SOC, upper and lower limits of SOC, and final-state SOC constraints of energy storage.

[0048]

[0049]

[0050]

[0051] in, , These represent the discharge power and charging power of the energy storage unit during time period t, respectively. , These are the maximum charging and discharging power and the maximum stored capacity of the energy storage unit, respectively. The energy storage unit is in its state of charge. To solve for the time step size; , These represent the minimum and maximum states of charge, respectively. , These are discharge and charge efficiency, respectively. Self-discharge rate; The final state of energy storage in the clearing result is the SOC. This represents the final-state SOC constraint value of the energy storage obtained by clearing the upper-level model.

[0052] S3. Divide the long-term simulation into M solution cycles. Using the system initial value or the final state of the clearing result of the previous solution cycle as the initial state, use a global simplified clearing model for prediction and optimization to obtain the optimized SOC constraint values ​​for the final state of energy storage in each cycle. Use only the SOC constraint value of the final state of energy storage in the current cycle to complete the electricity spot market clearing for the current solution cycle. Update the system state with the clearing result of the current cycle, repeat the global simplified clearing and detailed clearing, and execute the clearing process for subsequent cycles in a rolling manner.

[0053] The tiered rolling clearing process achieves long-term simulated clearing through a "prediction-execution-update" rolling mechanism. The specific steps are as follows: S301. Divide the long simulation into M solution cycles. The cycle length is determined by taking into account the solution speed and the operation cycle characteristics of energy storage. This forms a three-level time structure of "total simulation time, M solution cycles, and each cycle contains multiple 15-minute time intervals".

[0054] S302, Upper-layer prediction optimization.

[0055] The solution period of the upper-level global simplified clearing model changes dynamically. The first solution period covers the entire simulation duration, from the initial time to the end of the Mth solution cycle, with the initial value being the global initial state of the system. Subsequently, after each solution cycle completes spot market clearing, the starting point of the global simplified clearing model rolls back to the initial time of the current solution cycle, while the ending point remains unchanged; the initial state is the final state of the clearing result of the previous solution cycle.

[0056] Solving this upper-level optimization problem using a globally simplified clearing model yields the optimized SOC values ​​for the final state of energy storage in each cycle, as follows: , k is the number of remaining solution cycles in the current global simplified clearing model, which decreases gradually as the cycle continues. Only the SOC optimization value at the final state of energy storage in the current solution cycle is extracted as the constraint value. Used for lower-level execution. The obtained final-state SOC constraint value of the energy storage. Configure it using the following method: A simplified time scale is adopted, and the simulation period is set to 1 hour or several hours as needed. The data for each period is the average of the corresponding spot market 15-minute interval data.

[0057] The first solution period covers the entire simulation duration, from the initial moment to the end of the Mth solution cycle, with the initial value being the global initial state of the system. After each subsequent solution cycle of spot market clearing, the starting point of the global simplified clearing model rolls back to the initial moment of the current solution cycle, while the ending point remains unchanged; the initial state is the final state of the clearing result from the previous solution cycle.

[0058] The solution clearing is completed using a globally simplified clearing model. The clearing results include the optimized SOC values ​​for each time period. The optimized SOC value at the final state of the current solution period is extracted as the constraint value. Used for execution at the lower level.

[0059] S303, Detailed clearing of the spot market during the lower-level solution cycle.

[0060] The initial value for the first solution cycle is the system's global initial state, and the initial values ​​for subsequent solution cycles are the final state of the cleared results from the previous solution cycle. (This is used to obtain...) The final-state SOC constraint value of energy storage is used as the solution period. The clearing is completed using a detailed spot market clearing model to obtain detailed clearing results for each 15-minute period, including the output of thermal power units, renewable energy output, energy storage charging / discharging power, energy storage SOC, and nodal price for each period.

[0061] S304. Update the system state with the current cycle's final state, return to step 302, and repeat the upper-level prediction and lower-level detailed clearing until all M cycles have been cleared.

[0062] Example 1 This embodiment illustrates the clearing process, and the rolling clearing principle is as follows: Figure 2 As shown: In one example of this method, for a 30-day spot market simulated clearing, the initial state of the system is known, including the start-up and shutdown status and output of each unit at the initial moment, the initial SOC of energy storage is 50%, the relevant parameters of unit operation constraints and network constraints, the price curves of each type of unit every 15 minutes, the renewable energy output forecast, and the load forecast. It is required that the energy storage SOC return to 50% at the end of 30 days.

[0063] The simulation is divided into M=5 solution cycles, each lasting 6 days. In the global simplified clearing model, the time period is 1 hour, and the data for each time period is the average of the corresponding 15-minute intervals. In the spot market detailed clearing model, the time period is 15 minutes.

[0064] First round of execution: Using initial system parameters, the final-state SOC constraint of the globally simplified clearing model is 50%. The clearing problem, consisting of 5 cycles from day 1 to 30, is solved to obtain the optimized SOC values ​​at 24:00 on days 6, 12, 18, 24, and 30. The SOC value at 24:00 on day 6 is extracted as the final-state SOC constraint value for the first solution cycle. The system initial parameters and the SOC constraint values ​​are used. The system completes the clearing of the detailed spot market model for days 1-6, obtaining the simulated clearing results for each period. The final state on day 6 is used to update the system state as the initial value for the next round of solving.

[0065] Second round of execution: Using the updated state as the initial solution value, the final-state SOC constraint of the global simplified clearing model remains at 50%. The clearing problem for 7-30 days (including 4 cycles) is solved to obtain the optimized SOC values ​​at 24:00 on days 12, 18, 24, and 30. The SOC constraint value at 24:00 on day 12 is used as the final-state SOC constraint value for the second solution cycle. Using the updated initial state and the constraint value, complete a detailed spot market clearing process over 7-12 days and update the system status.

[0066] Following the logic described above, the cycle clearing is completed sequentially in 13-18 days, 19-24 days, and 25-30 days, ultimately achieving a 30-day simulation with the energy storage final-state SOC returning to 50%.

[0067] Example 3 The effects of the clearing method of the present invention will be illustrated below with specific examples.

[0068] Analysis and calculations were performed using a modified version of the IEEE standard 30-node case 30. The system network structure is identical to IEEE case 30. Thermal power generating units are located at nodes 1, 5, 8, and 13, with a total installed capacity of 2600MW; wind power generating units are located at node 2, with an installed capacity of 1000MW; photovoltaic units are located at node 11, with an installed capacity of 1000MW; and energy storage units are located at node 2, with a rated installed capacity of 200MW, a storage duration of 8 hours, and a charge / discharge efficiency of 90%. The load is distributed across nodes 2, 3, 4, 5, 7, 8, 10, 12, 17, 18, 21, 20, 26, and 30. Load data and renewable energy output forecast data were referenced from the actual situation in a Chinese province in June, and scaled according to their proportional relationship with the IEEE standard case. The total system load curve for that month is shown below. Figure 3 As shown.

[0069] To address the aforementioned issue of prolonged spot market clearing, a detailed global spot market clearing result over a long period (due to its long solution time) is used as a reference. Considering the 7-day periodicity of the actual data load, the 30-day spot market simulation is divided into 5 solution periods: the first 4 periods are 7 days each, and the last period is 2 days. A conventional periodic clearing method with equal initial and final state of charge (SOC) for each period and a hierarchical model-controlled prediction clearing method are used respectively. The clearing results are shown in Table 1.

[0070] Under the conventional periodic clearing method, the system's electricity purchase cost is the highest (371.5 million yuan), an increase of 610,000 yuan compared to the global optimum; while the energy storage unit profit is the lowest (1.123 million yuan), a decrease of 93,000 yuan compared to the global optimum. The main reason is that the constraint of equal state of charge (SOC) at the beginning and end of each cycle is too rigid, preventing energy storage from flexibly charging and discharging according to cross-cycle price trends. It forces a complete charge-discharge cycle within each cycle, even if price differences are not economically viable at certain times. This model not only limits the cross-cycle optimization capability of energy storage but also leads to inefficient energy conversion, failing to fully utilize the regulatory role of energy storage units and driving up the total system cost. The clearing method based on hierarchical model predictive control effectively approximates the global optimum. This method is significantly superior to the conventional periodic method, reducing the system's electricity purchase cost to 371.1 million yuan, while the energy storage unit profit of 1.193 million yuan is also higher than the conventional periodic clearing method.

[0071] Table 1. Comparison of clearing statistics under three clearing methods

[0072] Furthermore, the electricity price of system node 2 under conventional periodic clearing and the hierarchical model predictive control clearing method described in this invention is as follows: Figure 4 As shown, the electricity price for system node 2 on the last two days is as follows: Figure 5 As shown, the system node electricity prices are roughly the same under both clearing methods. However, during certain periods, the system node electricity price under the stratified model prediction and control clearing method is lower than that under the conventional periodic clearing method, resulting in lower system electricity purchase costs.

[0073] Energy storage SOC changes under conventional periodic clearing and the hierarchical model predictive control clearing method described in this invention are as follows: Figure 6 As shown, for conventional periodic clearing methods, the constraint that the state of charge (SOC) of energy storage is equal at the beginning and end of the solution cycle near the end makes it difficult for the charging and discharging behavior of energy storage to follow price signals, thus limiting the cross-cycle optimization capability of energy storage. The hierarchical model predictive control clearing method, because the final SOC value of each solution cycle is optimized, can better follow price signals and leverage the regulatory role of energy storage units.

[0074] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A simulated clearing method for the energy storage electricity spot market based on hierarchical model control and prediction, characterized in that, Includes the following steps: S1. Establish a simplified global clearing model for the electricity spot market that includes energy storage, as an upper-level predictive control model; S2. Establish a detailed clearing model for the electricity spot market that includes energy storage, as the underlying execution model; S3. Divide the long-term simulation into M solution cycles. In each solution cycle, first use global simplified clearing optimization to obtain the remaining energy storage final state SOC constraint values ​​for each cycle. Then, based on the energy storage final state SOC constraint values ​​for this cycle, use the detailed clearing model to complete the market simulation clearing for this cycle. Finally, update the system state according to the current clearing results and roll forward to the next cycle until all cycles are cleared.

2. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control and prediction as described in claim 1, characterized in that, In step S1, the key constraints retained in the global simplified clearing model of the electricity spot market include system power balance constraints, key section transmission constraints, thermal power unit operation constraints, energy storage operation constraints, and renewable energy output constraints.

3. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 2, characterized in that, In the global simplified clearing model, an aggregated time scale is used for simplification. The simulation period is set to 1 hour or several hours as needed, and the data for each period is the average of the corresponding spot market 15-minute interval data.

4. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 3, characterized in that, The mathematical model for the global simplified clearing model of the electricity spot market is as follows: The objective function is to minimize the total operating cost of the system. Where N is the number of generating units, and T0 is the number of time periods considered. Let be the operating cost of unit i in time period t, which is a piecewise linear function relating the unit's declared output ranges and corresponding energy prices. This represents the net output of unit i during time period t; System power balance constraints: in The total system load during time period t; Energy storage operation constraints include upper and lower limits of energy storage charge / discharge power, dynamic evolution constraints of SOC, upper and lower limits of SOC, and final-state SOC constraints of energy storage. in, , These represent the discharge power and charging power of the energy storage unit during time period t, respectively. , These are the maximum charging and discharging power and the maximum stored capacity of the energy storage unit, respectively. The energy storage unit is in its state of charge. This represents the simulation time step size; , These represent the minimum and maximum states of charge, respectively. , These are discharge and charge efficiency, respectively. Self-discharge rate; The final-state SOC constraint value for energy storage in the global simulation; Conventional constraints in the spot market, such as key section power flow constraints, thermal power unit operation constraints, and renewable energy output constraints: in, The planned power of critical section j in time period t. Let i be the active power output of thermal power unit i during time period t. For the active power output of renewable energy unit i in time period t, The upper limit of the power flow at the critical section j. , These are the lower and upper limits of the output of thermal power unit i, respectively. Let be the ramp rate of thermal power unit i. This represents the predicted output value of renewable energy unit i during time period t.

5. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 4, characterized in that, In step S2, the detailed clearing model of the electricity spot market with energy storage includes the safety-constrained unit combination model (SCUC) and the safety-constrained economic dispatch model (SCED) of the spot market. First, the start-up and shutdown status and output boundary of conventional units in each period are optimized and determined by the safety-constrained unit combination model (SCUC) of the spot market. Then, the output allocation of each unit is refined and optimized by the safety-constrained economic dispatch model (SCED) under the premise of fixed start-up and shutdown status, so as to ensure that the network security constraints and economic efficiency requirements are met. The nodal pricing mechanism is used to calculate the clearing price of each node.

6. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 5, characterized in that, Energy storage operation constraints include upper and lower limits of energy storage charge / discharge power, dynamic evolution constraints of SOC, upper and lower limits of SOC, and final-state SOC constraints of energy storage. in, , These represent the discharge power and charging power of the energy storage unit during time period t, respectively. , These are the maximum charging and discharging power and the maximum stored capacity of the energy storage unit, respectively. The energy storage unit is in its state of charge. To solve for the time step size; , These represent the minimum and maximum states of charge, respectively. , These are discharge and charge efficiency, respectively. Self-discharge rate; The final state of energy storage in the clearing result is the SOC. This represents the final-state SOC constraint value of the energy storage obtained by clearing the upper-level model.

7. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control and prediction according to claim 6, characterized in that, In step S3, the tiered rolling clearing process achieves long-term simulated clearing through a "prediction-execution-update" rolling mechanism. The specific steps are as follows: S301: Divide the long simulation into M solution cycles. The cycle length is determined by taking into account the solution speed and the operation cycle characteristics of energy storage. This forms a three-level time structure of "total simulation time, M solution cycles, and each cycle containing multiple 15-minute time intervals". S302: Upper-level prediction optimization. The solution period of the upper-level global simplified clearing model changes dynamically. The first solution period covers the entire simulation duration. The initial value is the global initial state of the system. After each solution cycle of spot market clearing is completed, the starting point of the global simplified clearing model rolls to the initial time of the current solution cycle, while the ending point remains unchanged. The initial state is the final state of the clearing result of the previous solution cycle. S303: Detailed clearing of the spot market in the next solution cycle. The initial value for the first solution cycle is the global initial state of the system, and the initial values ​​for subsequent solution cycles are the final state of the clearing result of the previous solution cycle, to obtain... The final state SOC constraint value of the energy storage as the solution period is cleared using a detailed clearing model of the spot market to obtain detailed clearing results for each 15-minute period. S304: Update the system state with the current cycle's final state, return to step 302, and repeat the upper-level prediction and lower-level detailed clearing until all M cycles have been cleared.

8. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 7, characterized in that, In step S302, the upper-level optimization problem is solved using a global simplified clearing model to obtain the optimized SOC values ​​for the final state of energy storage in each cycle, as follows: k represents the number of remaining solution cycles in the current global simplified clearing model, which decreases gradually as the model rolls on. Only the SOC optimization value at the final state of the energy storage in the current solution cycle is extracted as the constraint value. Used for execution at the lower level.

9. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 8, characterized in that, The final state of energy storage SOC constraint value The following settings are used: The first solution period covers the entire simulation duration, from the initial time to the end of the Mth solution cycle. The initial value is the global initial state of the system. After each subsequent solution cycle, the starting point of the global simplified clearing model is rolled to the initial time of the current solution cycle, while the ending point remains unchanged. The initial state is the final state of the clearing result of the previous solution cycle. The solution clearing is completed using a globally simplified clearing model. The clearing results include the optimized SOC values ​​for each time period. The optimized SOC value at the final state of the current solution period is extracted as the constraint value. Used for execution at the lower level.

10. The method for simulated clearing of the energy storage electricity spot market based on hierarchical model control prediction according to claim 9, characterized in that, In S303, the detailed clearing results include the output of thermal power units, renewable energy output, energy storage charging / discharging power, energy storage SOC, and nodal tariff for each time period.

Images