Real-time scheduling method and device for cross-regional transmission system of new energy base

Abstract

The invention discloses a real-time scheduling method and device for a new energy base cross-regional transmission system, and belongs to the technical field of power system control, and the real-time scheduling method comprises the steps: decoupling a full-time electric quantity constraint in a new energy power generation base cross-regional transmission system scheduling model, and reconstructing the model into a Markov decision process; the transmission power and the energy state of energy storage in the previous time period of the extra-high voltage direct current tie line are selected to fit the influence of the current decision and the residual electric quantity space of the system on the future optimal operation cost, and the high-precision approximation of the optimal operation cost function can be realized; carrying out piecewise linear representation on the optimal operation cost function and updating and iterating the slope of the optimal operation cost function; and finally, solving a scheduling decision of the current time period based on the updated slope, and sequentially solving a Markov decision process to obtain an optimal real-time scheduling strategy of the system. The accurate optimal operation cost function can successfully deal with real-time uncertainty, and the reliability and optimality of a real-time scheduling scheme are ensured.

Description

Technical Field

[0001] This invention belongs to the field of power system control technology, and more specifically, relates to a real-time scheduling method and device for a cross-regional transmission system of a new energy base. Background Technology

[0002] The uncertainties of large-scale renewable energy generation pose significant challenges to the reliable power supply and efficient renewable energy consumption of inter-regional transmission systems. Currently, large-scale renewable energy power generation bases have limited energy storage capacity and insufficient regulation capabilities, necessitating the introduction of new flexible resources to ensure the economical and safe operation of inter-regional transmission systems. By rationally optimizing the transmission power of ultra-high-voltage direct current (UHVDC) interconnection lines, their flexible adjustment potential can be fully utilized, effectively improving renewable energy consumption levels, eliminating potential power deficits, and significantly enhancing the flexibility of inter-regional transmission systems.

[0003] However, to ensure the reliable operation of DC converters and reduce their lifespan losses, the transmission power of UHVDC tie lines cannot be frequently and continuously adjusted. Furthermore, the daily cumulative transmission power of the tie lines must be executed according to the day-ahead inter-regional trading contract. At the same time, the real-time uncertainty of the loads in the wind power, photovoltaic, and receiving systems of the sending-end system in the inter-regional transmission also poses challenges to the accuracy and efficiency of system scheduling decisions.

[0004] Therefore, further in-depth research is needed on how to take into account the uncertainties of the system, explore the limited flexible adjustment capabilities of UHVDC interconnection lines, and provide a real-time optimized scheduling scheme for the inter-regional transmission system of large-scale new energy power generation bases. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a real-time scheduling method and device for a cross-regional transmission system of new energy bases. Its purpose is to solve the technical problem that the real-time uncertainty of the load in the cross-regional transmission sending system poses challenges to the accuracy and efficiency of system scheduling decisions.

[0006] To achieve the above objectives, according to one aspect of the present invention, a real-time scheduling method for a cross-regional transmission system of a new energy base is provided, comprising: S1: For the initial scheduling model of the inter-regional transmission system of the new energy power generation base including the UHVDC interconnection line, the full-time constraints in the initial scheduling model are decoupled into single-time constraints to obtain the target scheduling model, and then the target scheduling model is reconstructed into a Markov decision process. S2: Based on the Markov decision process, the transmission power and energy state of the UHVDC interconnection line in the previous period are selected to fit the initial optimal operating cost function of the inter-regional transmission system of the new energy power generation base from the current period to the end of the scheduling. S3: The target optimal operating cost function is obtained by performing piecewise linear characterization on the portion of the UHVDC interconnection line corresponding to the previous period transmission power and the portion of the energy storage corresponding to the energy state in the optimal operating cost function. S4: Iteratively update the two piecewise linear approximation functions in the target optimal operating cost function until training is complete; S5: Solve the Markov decision process using two trained piecewise linear value functions to obtain the optimal real-time scheduling strategy of the system, thereby completing real-time scheduling.

[0007] Furthermore, the single-time period constraint includes: daily cross-regional power transmission constraint and energy balance constraint for the first and last periods of energy storage.

[0008] Furthermore, the daily cross-regional power transmission constraint includes: ; ; ; ; ;

[0009] ; ; in, , These are, respectively, under the assumption that the constant power operation constraint of the UHVDC interconnection line is not considered, its current time period. The upper and lower limits of the cumulative transmitted power; This represents the percentage deviation between the daily inter-regional power transmission volume of the UHVDC interconnection line and the day-ahead inter-regional trading contract. For the total scheduling period The last period in, This refers to the day-ahead inter-regional trading contract electricity volume for UHVDC interconnection lines. For the constant power operation time of the UHVDC interconnection line, For any natural number, The interval between adjacent time periods; , These are the upper and lower limits of the transmission power of the UHVDC interconnection line (DC), respectively. , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable, respectively. , These are the upper and lower limits of the transmission power within the set of power-adjustable time periods when considering the constant power operation constraint of the UHVDC interconnection line; For the power transmitted in the previous period of the UHVDC interconnection line, , These refer to the maximum uphill and downhill climbing capabilities of the UHVDC interconnection line (DC). , These are, respectively, the conditions under which the DC constant power operation constraint of the UHVDC interconnection line is not considered, during the latter power adjustable period. The upper and lower limits of the cumulative transmitted power; , These are the upper and lower limits of the cumulative transmitted power during time period t, respectively, when considering the constant power operation constraint of the UHVDC interconnection line. , These are the DC lines of the UHVDC interconnection. , The cumulative amount of electricity transmitted during the time period.

[0010] Furthermore, the energy balance constraint during the first and last periods of energy storage includes: ; ; ; in, , These represent the upper and lower limits of the energy state of stored energy e during time period t. , These represent the maximum and minimum energy states of energy storage e, respectively. For the current time period, For the total scheduling period The last period in; , These are the charging and discharging efficiencies of energy storage e, respectively; , These are the maximum charging and discharging power of energy storage e, respectively; The energy state of energy storage e during the initial period of the total dispatch domain; Let e ​​be the energy state of the stored energy during time period t.

[0011] Furthermore, the target optimal operating cost function is expressed as: ; in, The optimal operating cost function for the current time period t. for and The corresponding operating cost of the system during time period t; This refers to the number of segments representing the power transmitted in the previous period of the UHVDC interconnection line (DC). For the current time period of the UHVDC interconnection line (dc) The slope of the transmitted power in the k-th segment of the piecewise linear approximation function, corresponding to the previous time period. The slope of the k-th segment in the piecewise linear approximation function of the transmitted power is expressed as: ; This refers to the estimated sampled values ​​of the slope of the linear piecewise function corresponding to the DC transmission power of the UHVDC interconnection line in the previous time period, including the forward sampled estimates. and negative sampling estimate , Update the step size for the slope; For the UHVDC interconnection line DC in the period preceding the kth segment Transmission power value; Let e ​​be the number of segments representing the energy state of the stored energy. Let be the slope of the l-th segment of the piecewise linear approximation function for the energy state of stored energy e; Let e ​​be the energy state value of the energy storage in segment l.

[0012] Furthermore, during the current time period of the UHVDC interconnection line (dc) satisfy When the transmission power is in the piecewise linear approximation function, the slope of the k-th segment is expressed as: forward sampled estimate. and negative sampling estimate Represented as: ;\ indicates exclusion; when the current time period of the UHVDC tie line DC. satisfy When the transmission power is in the piecewise linear approximation function, the slope of the k-th segment is expressed as: forward sampled estimate. and negative sampling estimate Represented as: ; in, For the total scheduling period The last period in, , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable. The interval between adjacent time periods. The constant power operating time of the UHVDC interconnection line; , These represent the positive and negative marginal contributions of the transmission power of the UHVDC interconnection line in the previous period to the optimal operating cost, respectively. , These are the positive and negative marginal flows of the UHVDC interconnection line's power transmission in the previous period, respectively. For the UHVDC interconnection line DC in the later power adjustable period The positive and negative slope estimates of the transmission power in the previous time period. For the UHVDC interconnection line DC in the later period The positive and negative slope estimates of the transmission power in the previous time period.

[0013] Furthermore, during the current time period of the UHVDC interconnection line (dc) satisfy: hour, .

[0014] According to another aspect of the present invention, a real-time scheduling device for a cross-regional transmission system of a new energy base is provided, comprising: The decoupling module is used to decouple the full-time constraints in the initial scheduling model of the inter-regional transmission system of new energy power generation bases that includes UHVDC interconnection lines into single-time constraints to obtain the target scheduling model, and then reconstruct the target scheduling model into a Markov decision process. The fitting module is used to select the transmission power and energy state of the UHVDC interconnection line in the previous period and fit the initial optimal operating cost function of the inter-regional transmission system of the new energy power generation base from the current period to the end of the scheduling, based on the Markov decision process. The characterization module is used to perform piecewise linear characterization of the transmission power corresponding to the previous period of the UHVDC tie line and the energy state corresponding to the energy storage in the optimal operating cost function to obtain the target optimal operating cost function. The update module is used to iteratively update the two piecewise linear approximation functions in the target optimal running cost function until training is completed; The solution module is used to solve the Markov decision process using two trained piecewise linear value functions to obtain the optimal real-time scheduling strategy of the system and thus complete the real-time scheduling.

[0015] According to another aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the real-time scheduling method described above.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the real-time scheduling method described above. In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: (1) This invention provides a real-time scheduling method for a cross-regional transmission system of a new energy base, decoupling the full-time power constraints in the scheduling model of a large-scale new energy power generation base cross-regional transmission system that takes into account the flexible adjustment capability of the UHVDC tie line, and reconstructing the model into a Markov decision process; selecting the transmission power and energy state of the UHVDC tie line in the previous time period to fit the impact of the current system decision and the remaining power space on the future optimal operating cost, which can achieve a high-precision approximation of the optimal operating cost function; the optimal operating cost function is characterized piecewise linearly, and its slope is updated iteratively according to the time period type; finally, the scheduling decision for the current time period is solved based on the actual system state and the updated slope, and the optimal real-time scheduling strategy of the system is obtained by sequentially solving the Markov decision process; the accurate optimal operating cost function can successfully cope with real-time uncertainty and ensure the reliability and optimality of the real-time scheduling scheme.

[0017] (2) The daily constraints in this scheme include: constant power operation constraints of UHVDC tie lines and daily cross-regional power transmission constraints, which can accurately describe the power adjustment characteristics and operation requirements of UHVDC tie lines, thereby fully exploring their flexibility adjustment potential.

[0018] (3) In this scheme, the all-time power constraint in the inter-regional transmission system scheduling model is decoupled into a single-time constraint. Among them, the daily inter-regional transmission power constraint of the UHVDC interconnection line is decoupled into... Decoupling of energy balance constraints during the first and last periods of energy storage It can meet the real-time scheduling process, thus realizing the feasibility of ensuring long-term rigid constraints in short-term scheduling.

[0019] (4) The optimal operating cost function in this scheme is expressed as: This ensures the accuracy of the optimal operating cost function fit and avoids the "curse of dimensionality"; it uses a piecewise linear function to approximate the function. Piecewise linear representation can accurately approximate the information of the value function and is easy to update, thus ensuring the accuracy of the real-time scheduling results and the speed of the solution.

[0020] (5) In this scheme, the slope of the UHVDC tie line transmission power corresponding to the piecewise approximation function in the previous period is updated in groups according to whether the period t is the UHVDC tie line power adjustable period. This can reflect the different impacts of the current state of the system and decisions on future operating costs in different types of periods, accurately embedding empirical knowledge into the slope of the piecewise linear function, and ensuring the correctness and effectiveness of the training process.

[0021] (6) This scheme reconstructs the scheduling model with constraint decoupling into a Markov decision process, where the set of state variables is represented as: Set of decision variables Represented as: It fully considers the impact of uncertainties in wind power, photovoltaics and load, and can model the sequential decision-making process of real-time scheduling, thereby realizing the solution of real-time scheduling decisions in an uncertain environment.

[0022] (7) This scheme fully considers the impact of wind power, photovoltaic and load randomness in the cross-regional transmission system. It generates a training scenario set based on the prediction information for day-ahead training, obtains the slope of the piecewise linear function with excellent performance, and applies it to the intraday real-time optimization to ensure the reliability and optimality of the real-time scheduling scheme. Attached Figure Description

[0023] Figure 1 This is a flowchart of the real-time scheduling method for the cross-regional transmission system of the new energy base provided in Example 1.

[0024] Figure 2 This is a structural diagram of the inter-regional transmission system for the large-scale new energy power generation base used in Example 1.

[0025] Figure 3a , Figure 3b and Figure 3c These are schematic diagrams of the day-ahead forecast curves and training datasets for wind power, photovoltaic power, and load under normal conditions, extreme high-generation conditions, extreme low-generation conditions, and extreme fluctuation conditions provided in Example 1.

[0026] Figure 4a , Figure 4b , Figure 4c and Figure 4d The figures are: a comparison of the real-time scheduling method of the new energy base inter-regional transmission system provided in Example 1 under normal conditions, a comparison of the effect under extreme high-yield conditions of new energy, a comparison of the effect under extreme low-yield conditions of new energy, and a comparison of the effect under extreme fluctuation conditions of new energy. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0028] Example 1 This embodiment provides a real-time scheduling method for a cross-regional transmission system in a new energy base, such as... Figure 1 As shown, it includes: S1-S5.

[0029] S1: For the initial scheduling model of the inter-regional transmission system of the new energy power generation base including the UHVDC interconnection line, the full-time constraints in the initial scheduling model are decoupled into single-time constraints to obtain the target scheduling model, and then the target scheduling model is reconstructed into a Markov decision process.

[0030] S2: Based on the Markov decision process, the transmission power and energy state of the UHVDC interconnection line in the previous period are selected to fit the initial optimal operating cost function of the inter-regional transmission system of the new energy power generation base from the current period to the end of the scheduling.

[0031] S3: The target optimal operating cost function is obtained by performing piecewise linear characterization on the portion corresponding to the transmission power of the UHVDC interconnection line in the previous time period and the portion corresponding to the energy state of the energy storage in the optimal operating cost function.

[0032] S4: Iteratively update the two piecewise linear approximation functions in the target optimal operating cost function until training is complete.

[0033] S5: Solve the Markov decision process using two trained piecewise linear value functions to obtain the optimal real-time scheduling strategy of the system, thereby completing real-time scheduling.

[0034] Furthermore, the single-time period constraint includes: daily cross-regional power transmission constraint and energy balance constraint for the first and last periods of energy storage.

[0035] Furthermore, the daily cross-regional power transmission constraint includes: ; ; ; ; ;

[0036] ; ; in, , These are, respectively, under the assumption that the constant power operation constraint of the UHVDC interconnection line is not considered, its current time period. The upper and lower limits of the cumulative transmitted power; This represents the percentage deviation between the daily inter-regional power transmission volume of the UHVDC interconnection line and the day-ahead inter-regional trading contract. For the total scheduling period The last period in, This refers to the day-ahead inter-regional trading contract electricity volume for UHVDC interconnection lines. For the constant power operation time of the UHVDC interconnection line, For any natural number, The interval between adjacent time periods; , These are the upper and lower limits of the transmission power of the UHVDC interconnection line (DC), respectively. , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable, respectively. , These are the upper and lower limits of the transmission power within the set of power-adjustable time periods when considering the constant power operation constraint of the UHVDC interconnection line; For the power transmitted in the previous period of the UHVDC interconnection line, , These refer to the maximum uphill and downhill climbing capabilities of the UHVDC interconnection line (DC). , These are, respectively, the conditions under which the DC constant power operation constraint of the UHVDC interconnection line is not considered, during the latter power adjustable period. The upper and lower limits of the cumulative transmitted power; , These are the upper and lower limits of the cumulative transmitted power during time period t, respectively, when considering the constant power operation constraint of the UHVDC interconnection line. , These are the DC lines of the UHVDC interconnection. , The cumulative amount of electricity transmitted during the time period.

[0037] Furthermore, the energy balance constraint during the first and last periods of energy storage includes: ; ; ; in, , These represent the upper and lower limits of the energy state of stored energy e during time period t. , These represent the maximum and minimum energy states of energy storage e, respectively. For the current time period, For the total scheduling period The last period in; , These are the charging and discharging efficiencies of energy storage e, respectively; , These are the maximum charging and discharging power of energy storage e, respectively; The energy state of energy storage e during the initial period of the total dispatch domain; Let e ​​be the energy state of the stored energy during time period t.

[0038] Furthermore, the target optimal operating cost function is expressed as: ; in, The optimal operating cost function for the current time period t. for and The corresponding operating cost of the system during time period t; This refers to the number of segments representing the power transmitted in the previous period of the UHVDC interconnection line (DC). For the current time period of the UHVDC interconnection line (dc) The slope of the transmitted power in the k-th segment of the piecewise linear approximation function, corresponding to the previous time period. The slope of the k-th segment in the piecewise linear approximation function of the transmitted power is expressed as: ; This refers to the estimated sampled values ​​of the slope of the linear piecewise function corresponding to the DC transmission power of the UHVDC interconnection line in the previous time period, including the forward sampled estimates. and negative sampling estimate , Update the step size for the slope; For the UHVDC interconnection line DC in the period preceding the kth segment Transmission power value; Let e ​​be the number of segments representing the energy state of the stored energy. Let be the slope of the l-th segment of the piecewise linear approximation function for the energy state of stored energy e; Let e ​​be the energy state value of the energy storage in segment l.

[0039] Furthermore, during the current time period of the UHVDC interconnection line (dc) satisfy When the transmission power is in the piecewise linear approximation function, the slope of the k-th segment is expressed as: forward sampled estimate. and negative sampling estimate Represented as: ;\ indicates exclusion; when the current time period of the UHVDC tie line DC. satisfy When the transmission power is in the piecewise linear approximation function, the slope of the k-th segment is expressed as: forward sampled estimate. and negative sampling estimate Represented as: ; in, For the total scheduling period The last period in, , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable. The interval between adjacent time periods. The constant power operating time of the UHVDC interconnection line; , These represent the positive and negative marginal contributions of the transmission power of the UHVDC interconnection line in the previous period to the optimal operating cost, respectively. , These are the positive and negative marginal flows of the UHVDC interconnection line's power transmission in the previous period, respectively. For the UHVDC interconnection line DC in the later power adjustable period The positive and negative slope estimates of the transmission power in the previous time period. For the UHVDC interconnection line DC in the later period The positive and negative slope estimates of the transmission power in the previous time period.

[0040] Furthermore, during the current time period of the UHVDC interconnection line (dc) satisfy: hour, .

[0041] To more clearly illustrate the real-time scheduling method for a cross-regional transmission system of a new energy power generation base proposed in this invention, the following embodiments are analyzed using a large-scale cross-regional transmission system of a new energy power generation base as an example. Figure 2 As shown. The system includes sending-end thermal power units, sending-end wind farms, sending-end photovoltaic farms, sending-end energy storage, ultra-high voltage DC interconnection lines, receiving-end thermal power units, and receiving-end loads.

[0042] In Example 1, S1 includes: collecting the technical parameters of each component in the inter-regional transmission system under study; each component of the model system includes an ultra-high voltage DC tie line, receiving-end load, receiving-end thermal power unit, sending-end thermal power unit, wind farm, photovoltaic power plant, and energy storage, thereby establishing a scheduling model for the inter-regional transmission system of a large-scale new energy power generation base that takes into account the flexible adjustment capability of the ultra-high voltage DC tie line.

[0043] Specifically, the technical parameters of each component include: 1) Wind power forecast for the sending-end system Photovoltaic forecast values and the load forecast of the receiving end system ; 2) Upper and lower limits of output of thermal power units at the sending and receiving ends , , , Maximum uphill and downhill gradient rates of thermal power units at the sending and receiving ends , , , Cost coefficient of thermal power units , , Peak-shaving cost coefficient of thermal power units , The load rate of thermal power units used to distinguish the type of peak-shaving task , .

[0044] 3) Upper and lower limits of transmission power of UHVDC interconnection lines , The maximum uphill and downhill gradient of the transmission power of the UHVDC interconnection line , UHVDC interconnection line constant power operation time The electricity volume of the UHVDC interconnection line cross-regional trading contract was recently announced. The acceptable percentage deviation between the daily inter-regional power transmission volume of UHVDC interconnection lines and the day-ahead inter-regional trading contracts. .

[0045] 4) Maximum charging and discharging power of energy storage , Upper and lower limits and initial values ​​of energy storage energy state , and Energy storage charging and discharging efficiency , .

[0046] 5) System wind curtailment penalty cost coefficient System light discard penalty cost coefficient The system's UHVDC transmission plan is not met, and the load shedding penalty factor is insufficient. .

[0047] The established scheduling model for the inter-regional transmission system of large-scale new energy power generation bases, taking into account the flexible adjustment capability of ultra-high voltage direct current transmission lines, has the following objective function: ; ; ; ; ; The constraints of the scheduling model for the inter-regional transmission system of large-scale new energy power generation bases, taking into account the flexible adjustment capabilities of UHVDC interconnection lines, include: upper and lower limits of UHVDC interconnection line transmission power, UHVDC interconnection line ramping constraints, UHVDC interconnection line constant power operation constraints, cumulative UHVDC interconnection line transmission power constraints, daily inter-regional UHVDC interconnection line transmission power constraints, upper and lower limits of thermal power unit output, thermal power unit ramping constraints, upper and lower limits of wind power, upper and lower limits of photovoltaic power, upper and lower limits of power constraints for UHVDC interconnection line transmission plans not meeting the upper and lower limits, upper and lower limits of load shedding power, energy storage energy state transition constraints, upper and lower limits of energy storage charging and discharging power, upper and lower limits of energy storage energy state, energy balance constraints during the first and last periods of energy storage, power balance constraints of the sending-end system, and power balance constraints of the receiving-end system.

[0048] In Example 1, the constraints of the scheduling model for the inter-regional transmission system of a large-scale new energy power generation base, taking into account the flexible adjustment capability of the UHVDC interconnection line, are expressed as follows: Upper and lower limits of transmission power constraints for UHVDC interconnection lines: ; UHVDC interconnection line ramping constraints: ; Constituent power operation constraints of UHVDC interconnection lines: ; Cumulative power transmission constraints for UHVDC interconnection lines: ; Daily cross-regional power transmission constraints for UHVDC interconnection lines: ; Upper and lower limits of thermal power unit output constraints: ; Thermal power unit ramping constraints: ; Wind power upper and lower limit constraints: ; Photovoltaic power upper and lower limit constraints: ; The UHVDC interconnection line transmission plan does not meet the upper and lower power limits constraints: ; Load shedding power upper and lower limit constraints: ; Energy storage state transition constraints: ; Upper and lower limits of energy storage charging and discharging power constraints: , ; Energy storage state of energy upper and lower limits constraints: ; Energy balance constraints during the first and last periods of energy storage: ; Power balance constraints of the sending-end system: ; Power balance constraints of the receiving end system: ; In Example 1, S1 further includes: decoupling the full-time power constraints in the cross-regional transmission system scheduling model into single-time constraints, and reconstructing the scheduling model after constraint decoupling into a Markov decision process. .

[0049] In Example 1, the decoupling method for all-time power constraints in the scheduling model is as follows: 1) The decoupling method for daily inter-regional power transmission constraints of UHVDC interconnection lines is as follows: ; ; ; ; ; ; ; ; in, , These are, respectively, under the assumption that the constant power operation constraint of the UHVDC interconnection line is not considered, its current time period. The upper and lower limits of the cumulative transmitted power; This represents the percentage deviation between the daily inter-regional power transmission volume of the UHVDC interconnection line and the day-ahead inter-regional trading contract. For the total scheduling period The last period in, This refers to the day-ahead inter-regional trading contract electricity volume for UHVDC interconnection lines. For the constant power operation time of the UHVDC interconnection line, For any natural number, The interval between adjacent time periods; , These are the upper and lower limits of the transmission power of the UHVDC interconnection line (DC), respectively. , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable, respectively. , These are the upper and lower limits of the transmission power within the set of power-adjustable time periods when considering the constant power operation constraint of the UHVDC interconnection line; For the power transmitted in the previous period of the UHVDC interconnection line, , These refer to the maximum uphill and downhill climbing capabilities of the UHVDC interconnection line (DC). , These represent the upper and lower limits of the cumulative transmitted power in the next power-adjustable period, assuming that the constant power operation constraint of the UHVDC interconnection line is not considered. , These are the upper and lower limits of the cumulative transmitted power during time period t, respectively, when considering the constant power operation constraint of the UHVDC interconnection line. , These are the DC lines of the UHVDC interconnection. , 1) The cumulative transmitted electricity during the time period. 2) The decoupling method for energy balance constraints at the beginning and end of the energy storage period is as follows: ; ; ; in, , These represent the upper and lower limits of the energy state of stored energy e during time period t. , These represent the maximum and minimum energy states of energy storage e, respectively. For the current time period, For the total scheduling period The last period in; , These are the charging and discharging efficiencies of energy storage e, respectively; , These are the maximum charging and discharging power of energy storage e, respectively; The energy state of energy storage e during the initial period of the total dispatch domain; Let e ​​be the energy state of the stored energy during time period t.

[0050] In Example 1, the scheduling model with decoupled constraints is reconstructed into a Markov decision process, as follows: 1) The state variables of a Markov decision process reflect the current state of the system. In this example, the set of state variables is: .

[0051] 2) The decision variables in a Markov decision process reflect the system's decisions at the current time point. In this example, the set of decision variables is: .

[0052] 3) The uncertainties in a Markov decision process reflect the uncertainty of the system. In this example, the set of uncertainties is as follows: .

[0053] in, , These represent the upper and lower limits of the output of the thermal power units at the sending and receiving ends during time period t; , These represent the charging and discharging power of energy storage e during time period t; Let w be the power of the wind farm that the system can obtain during time period t; Let p be the power of the photovoltaic electric field that the system can obtain during time period t; The load power of the system during time period t; Let w be the grid-connected power of the wind farm during time period t; Let p be the grid-connected power of the photovoltaic power plant during time period t; The load shedding power of the system during time period t; Let t be the predicted wind power output for time period t; Let be the predicted value of photovoltaic power during time period t; The predicted load of the receiving-end system during time period t; Let t be the prediction error of wind power output during time period t; Let be the prediction error of photovoltaic power in time period t; The prediction error of the system load during time period t is denoted as t.

[0054] 4) The state transition equation of a Markov decision process reflects the state changes of the system. In this example, the state transition equation is: , , , .

[0055] In Example 1, the piecewise linear function value approximation and slope update method proposed by S4 has the following specific steps: S41. Initialize the slopes of the previous period transmission power and the energy state of the stored energy in the piecewise linear approximation function of the UHVDC interconnection line, and let n=1. S42. Based on the predicted power output of each wind farm, photovoltaic farm, and load power of the system on the day-ahead, generate a set of uncertainties of the power system as a training dataset. S43, Order ; S44. Based on the Markov decision process, determine the state variables of the system in time period t, and solve for the decision variables of the system in time period t in the nth iteration. ; S45. Based on the piecewise linear function slope grouping update method distinguished by time period type, calculate the slope sampling estimates of the linear piecewise value function corresponding to the transmission power of the UHVDC tie line in the previous time period and the linear piecewise value function corresponding to the energy storage energy state, respectively, and use the formula... , Update the slope of the corresponding linear piecewise function; S46. Calculate the system based on the state transition equation. Status of the time period ; S47, Order Repeat steps S44-S46 until... ; S48, Order Repeat steps S42-S46 until... N is the preset maximum number of iterations; S49. Output the slope of the trained linear piecewise function.

[0056] In Example 1, the method for updating the slope of a piecewise linear function based on time period type is as follows: 1) The optimal operating cost function based on the piecewise linear function approximation is: ; ; ; ; in, Let be the optimal operating cost function for time period t. Let be the optimal operating cost function after the decision in time period t. This refers to the number of segments representing the power transmitted in the previous period of the UHVDC interconnection line (DC). The slope of the k-th segment in the piecewise linear approximation function of the power transmitted by the UHVDC tie line DC in the previous time period; The value of the transmitted power of the UHVDC interconnection line DC in the previous time period of the kth segment; Let e ​​be the number of segments representing the energy state of the stored energy. Let be the slope of the l-th segment of the piecewise linear approximation function for the energy state of stored energy e; Let e ​​be the energy state value of the energy storage in segment l.

[0057] 2) The method for updating the slope of a piecewise linear function based on time period type is as follows: If time period t is the power of the UHVDC interconnection line, the time period can be adjusted: ; ; If time period t is a period during which the power of the UHVDC interconnection line cannot be adjusted: ; ; For all time periods in the overall scheduling domain: ;

[0058] ;

[0059] ; ; in, , , , These represent the positive and negative marginal contributions of the UHVDC interconnection line's transmission power (dc) and energy storage (e) to the optimal operating cost, respectively. , , , These are the positive and negative marginal flows of the UHVDC interconnection line's transmission power and energy storage state in the previous period, respectively. , , , These are the positive and negative sampled estimates of the slope of the linear piecewise function corresponding to the transmission power of the UHVDC interconnection line dc in the previous period and the energy state of the energy storage e, respectively. Update the step size for the slope; \ indicates exclusion; , The transmission power of the UHVDC tie line DC in the period preceding time t is: , System operating costs during time period t; , The energy state of energy storage e in time period t is: , The system operating cost for time period t; k is the slope segment of the UHVDC interconnection line's transmission power in the previous time period; l is the slope segment of the energy storage state.

[0060] In Example 1, S5 includes: applying the trained piecewise linear function to real-time optimization scheduling to obtain the real-time optimal scheduling strategy for the cross-regional transmission system. The specific steps are as follows: S51, Order ; S52. Based on the true information of the uncertainty factors in time period t, determine the true state variables of the system in the current time period based on the Markov decision process; S53. Using the slope of the trained linear piecewise function, obtain the optimal scheduling decision for the system during time period t. ; S54, Order Repeat steps S52-S53 until... ; S55, Output the optimal real-time scheduling strategy of the system.

[0061] Simulation Results: To further illustrate the effectiveness of the real-time scheduling method for the inter-regional transmission system of new energy bases provided by this invention, this invention, in the following ways... Figure 2 A simulation comparison was performed on the large-scale inter-regional transmission system of the new energy power generation base shown, such as... Figure 3a , Figure 3b and Figure 3c The figures show the prediction errors of the system's stochastic factors under the assumptions of normal conditions, extreme high-volume conditions of new energy sources, extreme low-volume conditions of new energy sources, and extreme fluctuation conditions of new energy sources, respectively, assuming that these conditions follow a normal distribution. , .like Figure 4a , Figure 4b , Figure 4c and Figure 4d The figure shows a comparison of the performance of the real-time scheduling method (PD-ADP) for the inter-regional transmission system of new energy bases provided by this invention under normal conditions, extreme high-generation conditions, extreme low-generation conditions, and extreme fluctuation conditions of new energy, with the traditional ADP algorithm (TR-ADP) and the model predictive control algorithm (MPC) whose slope is not updated according to time period type. The horizontal axis represents the optimization error, and the vertical axis represents the probability of different optimization errors occurring. Figure 3a , Figure 3b and Figure 3cIt can be seen that the real-time scheduling method (PD-ADP) for inter-regional transmission systems in new energy bases proposed in this invention performs optimally. Under 200 test scenarios with normal conditions, the average optimization error between PD-ADP and the ideal example is 0.0099, significantly better than the 0.0749 and 0.0865 of TR-ADP and MPC algorithms, respectively. Under 200 test scenarios with extreme new energy generation conditions, the average optimization error between PD-ADP and the ideal example is 0.0178, significantly better than TR-ADP and MPC algorithms. The PC algorithm achieves an average optimization error of 0.0826 and 0.1025, respectively. In 200 test scenarios under extreme low-power conditions in new energy sources, the average optimization error of PD-ADP compared to the ideal example is 0.0210, significantly better than the TR-ADP and MPC algorithms' 0.1049 and 0.1138. In 200 test scenarios under extreme fluctuation conditions in new energy sources, the average optimization error of PD-ADP compared to the ideal example is 0.0180, significantly better than the TR-ADP and MPC algorithms' 0.0923 and 0.1055. Therefore, the real-time scheduling method for inter-regional transmission systems in new energy bases provided by this invention can fully leverage the flexible adjustment capabilities of UHVDC interconnection lines under uncertain environments, obtaining optimal real-time scheduling decisions for inter-regional transmission systems in large-scale new energy power generation bases, ensuring the economic efficiency and reliability of system operation.

[0062] Example 2 This embodiment provides a real-time scheduling device for a cross-regional transmission system of a new energy base, including: a decoupling module, a fitting module, a characterization module, an update module, and a solution module.

[0063] The decoupling module is used to decouple the full-time constraints in the initial scheduling model of the inter-regional transmission system of new energy power generation bases that includes UHVDC interconnection lines into single-time constraints to obtain the target scheduling model, and then reconstruct the target scheduling model into a Markov decision process.

[0064] The fitting module is used to fit the initial optimal operating cost function of the inter-regional transmission system of the new energy power generation base from the current period to the end of the scheduling period, based on the Markov decision process and the transmission power and energy state of the UHVDC interconnection line in the previous period.

[0065] The characterization module is used to perform piecewise linear characterization of the transmission power corresponding to the previous period of the UHVDC interconnection line and the energy state corresponding to the energy storage in the optimal operating cost function to obtain the target optimal operating cost function.

[0066] The update module is used to iteratively update the two piecewise linear approximation functions in the target optimal running cost function until training is completed.

[0067] The solution module is used to solve the Markov decision process using two trained piecewise linear value functions to obtain the optimal real-time scheduling strategy of the system and thus complete the real-time scheduling.

[0068] Example 3 The present invention also relates to an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0069] The electronic device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The memory can be used to store computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory.

[0070] Example 4 The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0071] Specifically, the memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0072] Example 5 This invention provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of the method described in the above embodiments of this invention.

[0073] The technical features of the embodiments described above can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. It should be noted that the terms "in one embodiment," "for example," and "again" in this invention are intended to illustrate the invention and are not intended to limit the invention.

[0074] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A real-time scheduling method for a cross-regional transmission system in a new energy base, characterized in that, include: S1: For the initial scheduling model of the inter-regional transmission system of the new energy power generation base including the UHVDC interconnection line, the full-time constraints in the initial scheduling model are decoupled into single-time constraints to obtain the target scheduling model, and then the target scheduling model is reconstructed into a Markov decision process. S2: Based on the Markov decision process, the transmission power and energy state of the UHVDC interconnection line in the previous period are selected to fit the initial optimal operating cost function of the inter-regional transmission system of the new energy power generation base from the current period to the end of the scheduling. S3: The target optimal operating cost function is obtained by performing piecewise linear characterization on the portion of the UHVDC interconnection line corresponding to the previous period transmission power and the portion of the energy storage corresponding to the energy state in the optimal operating cost function. S4: Iteratively update the two piecewise linear approximation functions in the target optimal operating cost function until training is complete; S5: Solve the Markov decision process using the two trained piecewise linear value functions to obtain the optimal real-time scheduling strategy of the system, thereby completing real-time scheduling.

2. The real-time scheduling method for the inter-regional transmission system of new energy bases as described in claim 1, characterized in that, The single-period constraints include: daily cross-regional power transmission constraints and energy balance constraints for the first and last periods of energy storage.

3. The real-time scheduling method for the inter-regional transmission system of new energy bases as described in claim 2, characterized in that, The daily cross-regional power transmission constraints include: ； ； ； ； ； ； ； in, , These are, respectively, under the assumption that the constant power operation constraint of the UHVDC interconnection line is not considered, its current time period. The upper and lower limits of the cumulative transmitted power; This represents the percentage deviation between the daily inter-regional power transmission volume of the UHVDC interconnection line and the day-ahead inter-regional trading contract. For the total scheduling period The last period in, This refers to the day-ahead inter-regional trading contract electricity volume for UHVDC interconnection lines. For the constant power operation time of the UHVDC interconnection line, For any natural number, The set of natural numbers, The interval between adjacent time periods; , These are the upper and lower limits of the transmission power of the UHVDC interconnection line (DC), respectively. , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable, respectively. , These are the upper and lower limits of the transmission power within the set of power-adjustable time periods when considering the constant power operation constraint of the UHVDC interconnection line; For the power transmitted in the previous period of the UHVDC interconnection line, , These refer to the maximum uphill and downhill climbing capabilities of the UHVDC interconnection line (DC). , These represent the upper and lower limits of the cumulative transmitted power in the next power-adjustable period, assuming that the constant power operation constraint of the UHVDC interconnection line is not considered. , These are the upper and lower limits of the cumulative transmitted power during time period t, respectively, when considering the constant power operation constraint of the UHVDC interconnection line. , These are the DC lines of the UHVDC interconnection. , The cumulative amount of electricity transmitted during the time period.

4. The new one as described in claim 2 The real-time scheduling method for inter-regional energy base transmission systems is characterized by, The energy balance constraints for the first and last periods of energy storage include: ； ； ； in, , These represent the upper and lower limits of the energy state of stored energy e during time period t. , These represent the maximum and minimum energy states of energy storage e, respectively. For the current time period, For the total scheduling period The last period in; , These are the charging and discharging efficiencies of energy storage e, respectively; , These are the maximum charging and discharging power of energy storage e, respectively; The energy state of energy storage e during the initial period of the total dispatch domain; Let e ​​be the energy state of the stored energy during time period t.

5. The real-time scheduling method for the inter-regional transmission system of new energy bases as described in claim 1, characterized in that, The target optimal operating cost function is expressed as: ； in, The optimal operating cost function for the current time period t. for and The corresponding operating cost of the system during time period t; This refers to the number of segments representing the power transmitted in the previous period of the UHVDC interconnection line (DC). For the current time period of the UHVDC interconnection line (dc) The slope of the transmitted power in the k-th segment of the piecewise linear approximation function, corresponding to the previous time period. The slope of the k-th segment in the piecewise linear approximation function of the transmitted power is expressed as: ; This refers to the estimated sampled values ​​of the slope of the linear piecewise function corresponding to the DC transmission power of the UHVDC interconnection line in the previous time period, including the forward sampled estimates. and negative sampling estimate , Update the step size for the slope; For the UHVDC interconnection line DC in the period preceding the kth segment Transmission power value; Let e ​​be the number of segments representing the energy state of the stored energy. Let be the slope of the l-th segment of the piecewise linear approximation function for the energy state of stored energy e; Let e ​​be the energy state value of the energy storage in segment l.

6. The real-time scheduling method for the inter-regional transmission system of new energy bases as described in claim 5, characterized in that, When the current time period of the UHVDC interconnection line DC satisfy When the transmission power is in the piecewise linear approximation function, the slope of the k-th segment is expressed as: forward sampled estimate. and negative sampling estimate Represented as: ;\ indicates exclusion; When the current time period of the UHVDC interconnection line DC satisfy When the transmission power is in the piecewise linear approximation function, the slope of the k-th segment is expressed as: forward sampled estimate. and negative sampling estimate Represented as: ; in, For the total scheduling period The last period in, , These are the sets of time periods where the transmission power of the UHVDC interconnection line (DC) is adjustable and non-adjustable. The interval between adjacent time periods. The constant power operating time of the UHVDC interconnection line; , These represent the positive and negative marginal contributions of the transmission power of the UHVDC interconnection line in the previous period to the optimal operating cost, respectively. , These are the positive and negative marginal flows of the UHVDC interconnection line's power transmission in the previous period, respectively. For the UHVDC interconnection line DC in the later power adjustable period The positive and negative slope estimates of the transmission power in the previous time period. For the UHVDC interconnection line DC in the later period The positive and negative slope estimates of the transmission power in the previous time period.

7. The real-time scheduling method for the inter-regional transmission system of new energy bases as described in claim 5, characterized in that, When the current time period of the UHVDC interconnection line DC satisfy hour, .

8. A real-time scheduling device for a cross-regional transmission system of a new energy base, characterized in that, include: The decoupling module is used to decouple the full-time constraints in the initial scheduling model of the inter-regional transmission system of new energy power generation bases that includes UHVDC interconnection lines into single-time constraints to obtain the target scheduling model, and then reconstruct the target scheduling model into a Markov decision process. The fitting module is used to select the transmission power and energy state of the UHVDC interconnection line in the previous period and fit the initial optimal operating cost function of the inter-regional transmission system of the new energy power generation base from the current period to the end of the scheduling, based on the Markov decision process. The characterization module is used to perform piecewise linear characterization of the transmission power corresponding to the previous period of the UHVDC tie line and the energy state corresponding to the energy storage in the optimal operating cost function to obtain the target optimal operating cost function. The update module is used to iteratively update the two piecewise linear approximation functions in the target optimal running cost function until training is completed; The solution module is used to solve the Markov decision process using two trained piecewise linear value functions to obtain the optimal real-time scheduling strategy of the system and thus complete the real-time scheduling.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images