Source-network collaborative planning method and system for spot market, medium and processor

Abstract

The invention discloses a source-network collaborative planning method and system for a spot market, a medium and a processor, relates to the technical field of power grid planning, and solves the problems of disjunction between traditional planning and dynamic supply and demand of the spot market, poor data quality and low solving efficiency. The method comprises the steps of collecting multi-dimensional operation parameters; a time sequence database is formed through cleaning, missing value processing, standardization and construction of a supply and demand balance coefficient K value time sequence curve; constructing a source-network collaborative planning model which takes the maximum social welfare of the power grid as a target and comprises power supply side, power grid side and flexibility resource constraints; and adopting an improved particle swarm optimization algorithm, and taking time sequence database data as an input solution model to obtain a planning scheme. According to the method, the supply and demand matching precision and data quality of the spot market are improved, the solving efficiency is optimized, the source-network collaborative dynamic adaptability is enhanced, and safe, economical and efficient operation of a power system is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of power grid planning technology, and in particular to a source-grid coordinated planning method, system, medium, and processor for the spot market. Background Technology

[0002] With the gradual establishment and improvement of the electricity spot market, the operation mode of the power system is transforming from traditional planned dispatch to market-based trading, and source-grid coordinated planning has become a core requirement to ensure the safe, economical and efficient operation of the power system.

[0003] In the current power system, the power generation side is characterized by a high proportion of new energy sources (photovoltaics, wind power, etc.) connected to the grid, significantly increasing the intermittency and volatility of their output. The grid side needs to address the contradiction between the increasing demand for cross-regional power transmission and the constraints of line transmission capacity. Meanwhile, user loads are developing towards diversification and flexibility, continuously increasing the difficulty of balancing supply and demand. However, traditional power generation and grid planning methods are mostly based on static historical data, failing to fully incorporate the dynamic supply and demand relationships in the spot market, leading to a disconnect between planning schemes and actual operational needs.

[0004] Specifically, existing planning methods have three main shortcomings: First, they lack quantitative indicators for supply-demand balance adapted to the spot market, making it difficult to accurately capture the dynamic matching relationship between real-time power output and load. Second, the data processing stage does not systematically clean and standardize multi-dimensional operating parameters (power output, grid flow, and user load), resulting in abnormal and missing data affecting planning accuracy. Third, the planning model solution often uses traditional optimization algorithms, which have slow convergence speeds and are prone to getting trapped in local optima, failing to efficiently output planning solutions that balance maximizing social welfare with satisfying system constraints. These problems lead to frequent supply-demand imbalances, line overloads, excessively high investment costs, or low operating efficiency in the power system under spot market conditions, hindering the healthy development of the power market and the sustainable operation of the power system.

[0005] Therefore, there is a need for a source-network collaborative planning method, system, medium, and processor for the spot market. Summary of the Invention

[0006] To address the frequent problems of supply-demand imbalance, line overload, excessive investment costs, or low operating efficiency in existing technologies, this invention provides a source-network collaborative planning method, system, medium, and processor for the spot market. This method improves the accuracy of supply-demand matching in the spot market, reduces the risk of supply-demand imbalance, optimizes planning solution efficiency and accuracy, achieves globally optimal planning, and enhances the dynamic adaptability of source-network collaboration. The specific technical solution is as follows: A source-network collaborative planning method for the spot market includes: S1: Collect multi-dimensional operating parameters; S2: Process and technically transform multi-dimensional operating parameters to obtain a time-series database; S3: Construct a source-network collaborative planning model; S4: An improved particle swarm optimization algorithm is used to solve the source-network collaborative planning model with standardized datasets and K-value time series curves in the time series database as inputs, and the source-network collaborative planning scheme is obtained.

[0007] Furthermore, in step S2, the processing and technical transformation of multi-dimensional operating parameters to obtain a time-series database includes the following steps: S21: Clean, handle missing values, and standardize the multi-dimensional operating parameters to obtain a standardized dataset; S22: Based on the supply and demand characteristics of the spot market, a supply and demand balance coefficient K is constructed as a technical indicator to obtain the time series curve of K value; S23: Store the processed standardized dataset and K-value time series curves into the time series database.

[0008] Furthermore, in step S21, the cleaning of multi-dimensional operating parameters includes using the 3σ criterion to remove abnormal data that exceeds the equipment's rated values. Specific steps include: Select the runtime parameters to be processed, and extract the corresponding original dataset, denoted as: ; in For the first The parameter values ​​at each sampling time point, This represents the total number of samples. Based on the node type to which the parameter belongs, retrieve the rated limits of the corresponding equipment, including: extracting the maximum output limit of the equipment for the power node. , minimum output lower limit Extract the rated transmission power of the power grid nodes. Rated current Extract the maximum load that the user-side node can withstand. ; The statistical threshold intervals are calculated based on the 3σ criterion, and the intervals are as follows: ; In the above formula, The mean of the dataset; The standard deviation of the dataset; The final determination range is constructed by superimposing the equipment rating values, including: The final normal range of power node parameters ; Final normal range of grid node parameters ; The final normal range of user-side node parameters ; Abnormal data that is not within the range is removed.

[0009] Furthermore, in step S22, the formula for calculating the K-value time series curve is as follows: ; In the formula, It represents the total real-time output of all power supply nodes in the power system; This represents the sum of real-time loads across all user-side nodes.

[0010] Furthermore, in step S3, constructing the source-network collaborative planning model includes the following steps: S31: The objective function of the source-grid collaborative planning model is to maximize the social welfare of the power grid system, as follows: Max W ; Where W represents the social welfare of the power grid system; Annualized investment cost; Annual operating and maintenance costs; For annual penalty costs; Benefits from power output; Revenue from power grid transmission services; S32: Based on the operating rules of the power system, three types of constraints are set for the source-grid collaborative planning model, including power source constraints, grid-side constraints, and flexibility resource constraints.

[0011] Furthermore, in step S4, the improved particle swarm optimization algorithm is used to solve the source-network collaborative planning model using the standardized dataset and K-value time series curves in the time series database as input, to obtain the source-network collaborative planning scheme, including the following steps: S41: Randomly generate several sets of initial planning schemes; S42: Calculate the objective function value W for each group of solutions, and select the solution with the largest W as the current optimal solution; S43: Update particle velocity and position based on the current optimal solution; S44: When the difference between the objective function values ​​of several consecutive iterations is less than the threshold, stop the iteration and output the global optimal solution; S45: Extract pure technical parameters from the global optimal solution to generate a source-network collaborative planning scheme.

[0012] Furthermore, the formula for updating the particle velocity and position is as follows: ; ; in, Let be the velocity of the i-th particle; ω be the inertial weight. , For learning factors; , A random number in the range [0,1]. is the individual optimal solution for the i-th particle; gbest is the global optimal solution; Let be the position of the i-th particle.

[0013] A source-network collaborative planning system for the spot market, applied to the aforementioned source-network collaborative planning method for the spot market, includes: The data acquisition module is used to collect multi-dimensional operating parameters. The processing module is used to process and technically transform multi-dimensional operating parameters to obtain a time-series database; The building module is used to construct the source-network collaborative planning model; The solution module is used to solve the source-network collaborative planning model by employing an improved particle swarm optimization algorithm, taking the standardized dataset and K-value time series curves in the time series database as input, and obtaining the source-network collaborative planning scheme.

[0014] A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the aforementioned source-network collaborative planning method for the spot market.

[0015] A processor for running a program, wherein the program executes the source-network collaborative planning method for the spot market described above.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. Improve the accuracy of supply and demand matching in the spot market and reduce the risk of supply and demand imbalance: The solution uses S1 multi-dimensional operational parameter collection (covering core parameters of power sources, grid, and users) and S2 technical transformation to build a time-series database and implicitly incorporates dynamic supply and demand correlation logic, breaking through the limitations of traditional planning's "static data dependence". Compared to existing technologies that lack specific quantitative indicators for supply and demand in the spot market, this solution can capture real-time fluctuations in power output, changes in grid transmission, and dynamic user load, providing accurate data support for subsequent planning and reducing market transaction risks and system adjustment costs caused by deviations in supply and demand predictions.

[0017] 2. Ensuring the quality of planning data and improving the reliability of planning schemes: The S2 stage's processing and technical transformation of multi-dimensional operating parameters specifically addresses the problems of incomplete data cleaning and lack of standardization in existing technologies. By systematically processing abnormal and missing data, a standardized dataset is formed, avoiding interference from invalid data on the planning model. This makes the input of the source-grid collaborative planning model constructed in the subsequent S3 stage more accurate, and the planning scheme more closely matches the actual operating conditions of the power system, reducing planning deviations caused by data errors.

[0018] 3. Optimizing planning efficiency and accuracy to achieve globally optimal planning: S4 employs an improved particle swarm optimization algorithm to solve the model. Compared to the slow convergence and susceptibility to local optima of traditional optimization algorithms in existing technologies, this algorithm can efficiently iterate based on standardized datasets in a time-series database to quickly locate the globally optimal solution. This ensures that the planning scheme meets various constraints such as those related to power generation, grid, and flexibility resources, while maximizing the social welfare of the power grid system. It effectively balances investment costs, operation and maintenance costs, and market benefits, avoiding the planning imbalance problems of "emphasizing costs over benefits" or "emphasizing benefits over constraints" in existing technologies.

[0019] 4. Enhance the dynamic adaptability of power generation and grid coordination to align with the market-oriented characteristics of the spot market: The solution, through a full-process design of "data acquisition - processing and transformation - model building - algorithm solving," integrates the real-time and dynamic nature of the spot market into all stages of planning, breaking through the bottleneck of traditional static planning being disconnected from dynamic market operation. The planning scheme can be dynamically adjusted according to changes in supply and demand in the spot market (such as real-time output and load fluctuations), ensuring that the allocation of power generation and grid resources always adapts to market trading needs. This reduces resource waste or system risks caused by the "one-time formulation and long-term difficulty in adjustment" of planning schemes in existing technologies, and improves the operational flexibility and economy of the power system in the spot market environment. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0021] Figure 1 This is a flowchart illustrating a source-network collaborative planning method for the spot market. Figure 2 This is a schematic diagram of a source-network collaborative planning system for the spot market. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] It should be understood that, when used in this application, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof.

[0024] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0025] It should also be further understood that the term “and / or” as used in this application refers to any combination of one or more of the associated listed items, as well as all possible combinations, and includes these combinations.

[0026] Example 1 like Figure 1 The diagram shows a flow chart of a source-network collaborative planning method for the spot market, which includes the following steps: S1: Collect multi-dimensional operating parameters.

[0027] S11: Selecting the data acquisition node.

[0028] For the power system covered by the spot market, three types of core data acquisition nodes are selected: Power nodes include thermal power plants, photovoltaic power plants, wind farms, etc., and the outgoing line end of each power node is selected as the parameter acquisition point.

[0029] The power grid nodes include the busbars, outgoing lines, and inter-provincial (inter-regional) tie lines of 220kV and above substations as data collection points.

[0030] User-side nodes include industrial, commercial, and residential users, with the incoming line end of the user's power distribution room selected as the data collection point.

[0031] S12: Deploy acquisition devices and acquire parameters at the acquisition nodes.

[0032] Power sensors are deployed at power nodes and connected to the power plant monitoring system via RS485 interface to collect real-time output values, maximum output upper limit, and minimum output lower limit at a frequency of ≥1 time / second.

[0033] Voltage / current sensors are deployed at grid nodes and connected to the substation SCADA system to collect and calculate node power flow data. The sensor accuracy is calibrated to an error of ≤0.5%.

[0034] Deploy load monitoring terminals at user-side nodes and collect real-time load values, load fluctuation frequency, and interruptible load ratio via LoRa wireless modules to ensure data sampling integrity ≥99.9%.

[0035] S13: Establish a transmission link between the data acquisition nodes and the central data center to upload data.

[0036] Edge computing devices are deployed in substations and key user areas. Communication connections are established between each data acquisition node device and the edge computing device. The edge computing device is then connected to the central data center via a 5G private network (bandwidth ≥ 100Mbps) to control data transmission latency ≤ 100ms and packet loss rate ≤ 0.1%.

[0037] S2: Process and technically transform multi-dimensional operating parameters to obtain a time-series database.

[0038] S21: Clean the multi-dimensional operating parameters, handle missing values, and standardize them to obtain a standardized dataset.

[0039] The central data center receives multi-dimensional operational parameters, including power output, grid parameters, and user load, which are then cleaned, have missing values ​​removed, and are standardized. Specifically, this includes: 1. Use the 3σ criterion to eliminate abnormal data that exceeds the equipment's rated values ​​(such as data where the power exceeds the power supply's maximum output).

[0040] 1.1 Select the operating parameters to be processed (such as real-time output values ​​of power generation nodes, power flow data of grid nodes, and real-time load values ​​of user-side nodes), and extract the corresponding original datasets, denoted as: ,in For the first The parameter values ​​at each sampling time point, This represents the total number of samples.

[0041] 1.2 Based on the node type to which the parameter belongs, retrieve the rated limit value of the corresponding device: Power Node (Thermal Power Plant / Photovoltaic / Wind Farm): Extracts the maximum output limit of the equipment. , minimum output lower limit ; Grid nodes (220kV and above substations, tie lines): Extract the rated transmission power of the lines. Rated current ; User-side nodes (industrial / commercial / residential users): Extract the maximum load that the user side can withstand. .

[0042] 1.3 Calculate the statistical threshold interval based on the 3σ criterion.

[0043] Calculate the mean of the dataset For the original dataset The formula for calculating the arithmetic mean is: in, This represents the average level of the parameter within the sampling period, reflecting the central trend of the data.

[0044] Calculate the standard deviation of the dataset Based on the mean The formula for calculating the dispersion of data is (using the sample standard deviation, with the denominator being...). To avoid bias): ;in, The larger the value, the more drastic the parameter fluctuations; conversely, the smaller the value, the more stable the parameter.

[0045] Determine the 3σ statistical threshold interval based on the characteristics of the normal distribution (approximately 99.73% of the data fall within this range). Within the range), calculate the statistically normal data interval: ; 1.4 Construct the final judgment range by superimposing the equipment rating values.

[0046] The 3σ statistical threshold interval is superimposed with the equipment rated value constraint, and the "intersection interval under the dual constraints" is taken as the final normal data determination interval. The specific rules are as follows: Power node parameters (such as real-time output) ): Final normal range It must meet the statistically normal fluctuation range, but must not exceed the equipment's own maximum / minimum rated output limit (to avoid equipment overload or inefficient operation).

[0047] Grid node parameters (such as line transmission power) ): Final normal range The power transmission capacity of the power grid cannot be negative (excluding negative values ​​caused by data acquisition errors), and it cannot exceed the rated transmission capacity of the line (to avoid overload and burnout of the line).

[0048] User-side node parameters (such as real-time load) ): Final normal range The user load cannot be negative (to exclude data collection errors) and cannot exceed the maximum load that the user side can withstand (to avoid tripping of user-side equipment).

[0049] 1.5 Remove outlier data that is not within the range.

[0050] Data point-by-data point determination: Traverse the original dataset Each data point in Determine whether it falls within the "final judgment interval": like Final judgment range: Data judged as normal will be retained for subsequent standardization processing. like Final judgment range: Data that is judged as abnormal (may be caused by equipment failure, acquisition error, or extreme operating conditions) is marked and removed.

[0051] Abnormal data recording: For excluded abnormal data, record it in the format of "Node type·Collection time-Parameter name-Abnormal value-Judgment basis" (e.g., "Power supply side-202405011000-Real-time output-850MW-Exceeding"). And exceed "), to ensure data traceability.

[0052] 2. Use linear interpolation to impute missing data, ensuring a missing data rate of ≤0.1%; Let the preceding known data point in the time series be . The next known data point is ,in: , The collection time for known data points (unit: seconds / minute, consistent with the data collection frequency); , These are known parameter values ​​for the corresponding time period (such as power output, load value, etc.). The collection time for missing data points is (satisfy The missing values ​​to be filled are: .

[0053] Then missing values The calculation formula is: .

[0054] 3. Standardize the data (map parameter values ​​to the [0,1] interval) to form a standardized dataset.

[0055] MinMax standardization (deviation standardization) is suitable for scenarios where the relative distribution characteristics of parameters need to be preserved. The formula is: ; The standardized value of the i-th data point (range [0,1]); This is the original processed value of the i-th data point (with anomaly removal and missing value imputation completed). It is the minimum value of the parameter dataset (taken from the lower limit of the "final normal judgment interval"); This is the maximum value of the parameter dataset (taken from the upper limit of the "final normal judgment interval").

[0056] S22: Based on the supply and demand characteristics of the spot market, a supply and demand balance coefficient K is constructed as a technical indicator, and a time series curve of the K value is obtained. The calculation formula is as follows: ; In the formula, The total real-time output of all power sources (thermal power plants, photovoltaic power plants, wind farms, etc.) in the power system (unit: MW); This represents the total real-time load of all user-side nodes (industrial, commercial, and residential users) (unit: MW).

[0057] Calculation frequency: The K value is updated every 1 minute to form a time series curve.

[0058] Physical meaning This indicates that there is a load gap in the system. This indicates that the power output is excessive. This indicates that supply and demand are in perfect balance.

[0059] S23: Store the processed standardized dataset and K-value time series curves to a time series database (such as InfluxDB), and label the data in the format of "node type-acquisition time-parameter name" (e.g., power supply side-202405011000-real-time output) to ensure data traceability and query response time ≤1 second.

[0060] S3: Construct a source-network collaborative planning model.

[0061] S31: Considering the planning model under a market environment, the objective function of the source-grid coordinated planning model is to maximize the social welfare of the power grid system, as follows: Max W ; Where W represents the social welfare of the power grid system (unit: 10,000 yuan). Annualized investment cost (spreading a one-time investment over the year); Annual operating and maintenance costs; This refers to the annual penalty costs incurred in the spot market due to supply and demand imbalances and failure to meet targets. The revenue from power output is calculated based on the actual power output of the power source and the value coefficient of the system output. The revenue from power grid transmission services is calculated based on the transmission power of the lines and the transmission service value coefficient.

[0062] ; in, For the first Installed capacity of power supply type (unit: MW); For the first Annual utilization hours of this type of power supply; The power output value coefficient (unit: RMB 10,000 / MW·h, set according to technical characteristics, environmental benefits, and reliability; for example, the coefficient for new energy is higher than that for thermal power due to its clean attributes).

[0063] ; in, For the first Transmission power per line (unit: MW); For the first Annual operating hours of each line; Value coefficient for power grid transmission (unit: RMB 10,000 / MW·h, set according to line voltage level, transmission efficiency, and network importance).

[0064] Furthermore, annualized investment costs ( The investment amortization covering power supply, power grid, and flexibility resources is calculated using the following formula: ; ; ; ; in, For power supply investment costs; This represents the total number of newly added power sources; For the first Installed capacity of power supply type (unit: MW); For the first Unit investment cost of power sources (unit: RMB 10,000 / MW, e.g., photovoltaic RMB 4 million / MW, wind power RMB 3.8 million / MW). ; Investment recovery coefficient The annual interest rate is For the economical lifespan of the power supply, it is usually taken as 20-25 years.

[0065] For power grid investment costs, This represents the total number of newly added / expanded lines; Let be the length of the j-th line (unit: km); The unit investment cost for the j-th line (unit: 10,000 yuan / km, e.g., 3 million yuan / km for a 220kV line and 8 million yuan / km for a 500kV line). The investment recovery factor is i (where i is the annual interest rate and T is the economic life of the line, usually taken as 30-40 years).

[0066] To reduce investment costs for flexible resources; This represents the total number of energy storage systems. For the first Rated capacity of each energy storage system (unit: MWh); For the first Unit investment cost of an energy storage system (unit: RMB 10,000 / MWh, such as RMB 15 million / MWh for lithium battery energy storage); The investment recovery factor ( The annual interest rate is The economic lifespan of energy storage is typically 10-15 years.

[0067] Furthermore, annual operating and maintenance costs ( Costs including routine equipment maintenance, energy consumption, and fuel are calculated using the following formula: ; ; ; ; For power supply operating costs; For thermal power plant operating costs; This refers to the annual utilization hours of thermal power units. Price per unit coal (unit: yuan / ton); Operating costs for new energy sources (photovoltaic / wind power); The unit operation and maintenance cost of new energy (unit: RMB 10,000 / MW·year, such as RMB 20,000 / MW·year for photovoltaic).

[0068] Furthermore, power grid operating costs The calculation formula is as follows: ; The transmission power of the j-th line (unit: MW); This refers to the annual operating hours of the line. The unit cost of line loss is RMB / kWh, based on the average spot market electricity price.

[0069] Furthermore, the operating costs of flexible resources The calculation formula is as follows: In the above formula, The unit operation and maintenance cost of energy storage (unit: RMB 10,000 / MWh·year, usually 2%-3% of the investment cost); The rated capacity of the kth energy storage system is expressed in MWh (megawatt-hours), reflecting the upper limit of the energy storage capacity of the energy storage device. The total number of energy storage systems involved in the calculation, i.e., the total number of energy storage devices planned / sets.

[0070] Furthermore, annual penalty costs To address the imbalance between supply and demand and the failure of technical indicators to meet standards in the spot market, the calculation formula is as follows: ; ; ; in, Punishment for supply and demand imbalance; The unit is the penalty coefficient for supply and demand imbalance (unit: 10,000 yuan); For the first Hourly supply and demand balance coefficient; 8760 represents the annual number of hours. Penalties for failing to meet targets; , These are SAIDI (System Average Outage Duration) and line loss rate ( The unit penalty coefficient; These are the target thresholds for SAIDI and line loss rate (e.g., SAIDI ≤ 5 minutes / year). .

[0071] S32: Based on the operating laws of the power system, three types of constraints are set for the source-grid coordinated planning model, specifically including: 1. Power supply side constraints.

[0072] 1.1 Output Constraints: This is the minimum output of the power supply. (Maximum power output) 1.2 Constraints for thermal power source measurement also include: Uphill constraint: The maximum output increase per unit time must not exceed the equipment's rated ramp rate, as shown in the formula: ; in, For the first Taiwan thermal power units Actual output at any given moment (unit: MW); For the first Rated uphill ramp rate of thermal power units (unit: MW / min, conventional thermal power is usually 296-596 rated capacity / min); The time step is measured in minutes (5-15 minutes in the spot market).

[0073] Thermal power plant downward ramp constraint: The maximum output decrease per unit time shall not exceed the equipment's rated ramp rate, as shown in the formula: ; in, For the first The rated downward ramp rate of a thermal power unit (unit: MW / min, usually slightly greater than the upward ramp rate, about 3%-6% of rated capacity / min).

[0074] Minimum operating time constraint: After the unit starts up, it must run continuously for at least a set time to avoid frequent start-stop damage to the equipment. The formula is: ; in, For the first Operating status of the thermal power unit (1 indicates operation, 0 indicates shutdown); For the unit from the last startup to Continuous runtime at any given time (in hours); This refers to the minimum continuous operating time of the unit (unit: h; conventional thermal power plants typically operate for 4-8 hours).

[0075] Minimum downtime constraint: After a unit is shut down, it must remain in a downtime state for at least a set period, calculated using the formula 20. ; in, For the unit since the last shutdown Continuous outage duration at any given time (in hours); This is the minimum continuous outage time for the unit (unit: h; for conventional thermal power plants, it is usually 2-4 hours).

[0076] 1.3 Constraints on the new energy power supply side also include: Uphill constraint: Limited by the maximum rate of change of natural resources, the formula is: ; in, The actual power output (in MW) of the i-th renewable energy power station (photovoltaic / wind power) at time t; The maximum upward ramp coefficient for new energy power plants (unit: 1 / min, photovoltaic is usually ≤0.05 / min, wind power is usually ≤0.03 / min). Rated installed capacity of new energy power plants (unit: MW).

[0077] Downhill constraint: The formula is: in, This is the maximum downward ramp coefficient for new energy power plants (unit: 1 / min, usually the same as or slightly larger than the upward ramp coefficient).

[0078] New energy power plants (photovoltaic / wind power) achieve rapid start-up and shutdown via inverters, with no minimum operation / outage time constraints. They only need to correlate output with status, as shown in the formula: ; in, This indicates the operating status of the renewable energy power station (1 indicates grid connection, 0 indicates grid disconnection); for The maximum generating power of a new energy power station at any given time (calculated from irradiance and wind speed, unit: MW).

[0079] 2. Grid-side constraints.

[0080] 2.1 Current constraint: 2.2 Power grid side line capacity constraints: in, The actual transmission power (in MW) of the j-th line at time t is calculated in real time from the power flow data of the power grid nodes in the "Multi-dimensional Operation Parameter Acquisition Steps". The rated transmission capacity (in MW) of line j is determined by the line type (e.g., 220kV line, 500kV line), referring to the State Grid design standards. For example, the rated transmission capacity of a 220kV line is typically 150-300MW, and that of a 500kV line is typically 750- .

[0081] 3. Flexibility resource constraints, specifically including: Energy storage power constraints: For energy storage charging and discharging power, For minimum charge and discharge power, This refers to the rated charge / discharge power.

[0082] Real-time charging Range constraints: ;in The minimum state of charge (usually 5%-10%, to avoid over-discharge and damage to the battery) This is the highest state of charge (usually 90%-95%, to avoid overcharging).

[0083] SOC rate of change constraint: ; The maximum permissible rate of change of SOC per unit time (e.g., ≤5% / min for lithium batteries); The time step is consistent with the time granularity of the spot market, 5-15 minutes.

[0084] Initial State of Charge (SOC) Constraint: At the start of the planning period, the SOC of the energy storage system must be within a preset range. (Usually, 30%-70% is taken to ensure initial adjustment capability).

[0085] Single charging duration constraint The duration of the k-th charge. The maximum duration of a single charge (determined by the rated energy storage capacity and maximum charging power) ).

[0086] Single discharge duration constraint: , Let be the duration of the k-th discharge. (Maximum discharge power).

[0087] Continuous charge / discharge interval constraint: A minimum interval must be maintained between two charges (or discharges). (For lithium batteries, ≥30 minutes, avoid frequent charging and discharging to prevent thermal runaway).

[0088] Daily charge / discharge cycles constraint: , This refers to the number of charge / discharge cycles per day (one charge + one discharge counts as one cycle). The settings are based on the type of energy storage (e.g., lithium batteries ≤ 2 times / day, vanadium redox flow batteries ≤ 4 times / day).

[0089] Annual charge / discharge cycles constraint: This refers to the cumulative number of charge-discharge cycles per year. The permissible average number of cycles per year over the lifespan of energy storage (e.g., lithium batteries) .

[0090] Deep charge / discharge cycle constraints: , For deep charge and discharge (SOC from ≤20% to ≥80%, or from ≥80% to ≥80%) The annual limit for the number of cycles is usually 30%-50% of the total number of cycles per year (to reduce the impact of deep cycling on battery life); a complete charge-discharge cycle must include both charging and discharging processes.

[0091] S4: An improved particle swarm optimization (PSO) algorithm is used to solve the source-network collaborative planning model with standardized datasets and K-value time series curves in the time series database as inputs, and the source-network collaborative planning scheme is obtained.

[0092] Furthermore, the parameters of the improved particle swarm optimization (PSO) algorithm are set as follows: Population size: 50 (50 planning schemes are generated in each iteration); Number of iterations: 100 (until the objective function value converges); Inertia weight (Balancing global and local search in the control algorithm); The objective function, constraints, and improved PSO algorithm are integrated to develop a model solver that supports parameter input and result visualization output.

[0093] Furthermore, the improved Particle Swarm Optimization (PSO) algorithm is used to solve the source-network collaborative planning model using a standardized dataset and K-value time series curves as input, resulting in a source-network collaborative planning scheme. This process includes the following iterative steps: S41: Randomly generate 50 initial planning schemes (including power supply configuration, grid topology, and energy storage parameters); S42: Calculate the objective function value W for each group of solutions, and select the solution with the largest W as the current optimal solution; S43: Update the particle velocity and position (planning parameters) based on the current optimal solution. The formula for updating velocity and position is: ; ; in, Let be the velocity of the i-th particle; ω be the inertial weight. , For learning factors; , A random number in the range [0,1]. is the individual optimal solution for the i-th particle; gbest is the global optimal solution; Let be the position of the i-th particle.

[0094] S44: When the difference in the objective function value after 10 consecutive iterations is ≤0.01%, stop the iteration and output the global optimal solution.

[0095] S45: Extract pure technical parameters from the global optimal solution to generate a source-network collaborative planning scheme, including: Power configuration: Add power type (e.g., photovoltaic / wind power), installed capacity, and installation location coordinates; Power grid planning: new / expanded line types, origin-end node numbers, and transmission capacity; Flexible resource allocation: energy storage system capacity, installation location, and charge / discharge power curves.

[0096] Beneficial effects: 1. Improve the accuracy of supply and demand matching in the spot market. The solution covers the collection of multi-dimensional operating parameters from the power supply, power grid, and user side, and constructs a time-series database including a supply-demand balance coefficient K to accurately capture the dynamic correlation between real-time output and load. Compared with existing technologies that lack dedicated quantitative indicators for supply and demand, this solution can reduce supply and demand forecasting errors, lower market transaction risks and system adjustment costs, and alleviate supply and demand imbalances.

[0097] 2. Ensure the quality of planning data and the reliability of the plan. By using the 3σ criterion combined with equipment ratings to eliminate outlier data, and then filling in missing values ​​with linear interpolation and standardizing the data, the shortcomings of existing technologies, such as incomplete data cleaning and lack of standardization, are resolved. The clean, standardized dataset allows for more accurate input to the planning model, ensuring the solution closely matches the actual operating conditions of the power system and reducing planning deviations caused by data errors.

[0098] 3. Optimize the efficiency of the programming solution and the global optimality. An improved particle swarm optimization algorithm is used to solve the model. Compared with traditional algorithms, which suffer from slow convergence and susceptibility to local optima, this algorithm can efficiently iterate based on time-series data and quickly locate the global optimum. While satisfying multiple constraints from the power source side, the grid side, and flexibility resources, it maximizes the social welfare of the power grid system, balances investment, operation and maintenance costs with market benefits, and avoids planning imbalances.

[0099] 4. Enhance the dynamic adaptability of source-network collaboration. The entire process incorporates the real-time and dynamic characteristics of the spot market, breaking through the bottleneck of the disconnect between traditional static planning and dynamic market operation. The planning scheme can be dynamically adjusted according to changes in supply and demand, ensuring that the allocation of power generation and grid resources always adapts to market trading needs, reducing resource waste or system risks, and improving the flexibility and economy of power system operation.

[0100] Example 2 like Figure 2 The diagram shows a source-network collaborative planning system for the spot market, applied to the aforementioned source-network collaborative planning method for the spot market, including: The data acquisition module is used to collect multi-dimensional operating parameters. The processing module is used to process and technically transform multi-dimensional operating parameters to obtain a time-series database; The building module is used to construct the source-network collaborative planning model; The solution module is used to solve the source-network collaborative planning model by employing an improved particle swarm optimization algorithm, taking the standardized dataset and K-value time series curves in the time series database as input, and obtaining the source-network collaborative planning scheme.

[0101] Example 3 A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the aforementioned source-network collaborative planning method for the spot market.

[0102] Example 4 A processor for running a program, wherein the program executes the source-network collaborative planning method for the spot market described above.

[0103] This application discloses a source-grid collaborative planning method, system, medium, and processor for the spot market, relating to the field of power grid planning technology. It addresses the problems of traditional planning being disconnected from the dynamic supply and demand of the spot market, poor data quality, and low solution efficiency. The method includes: collecting multi-dimensional operating parameters; cleaning, handling missing values, standardizing, and constructing time-series curves of the supply-demand balance coefficient K to form a time-series database; constructing a source-grid collaborative planning model with the objective of maximizing the social welfare of the power grid, including constraints on the power source side, the grid side, and flexibility resources; and using an improved particle swarm optimization algorithm, with the time-series database data as input, to solve the model and obtain the planning scheme. This application improves the accuracy and data quality of supply and demand matching in the spot market, optimizes solution efficiency, strengthens the dynamic adaptability of source-grid collaboration, and ensures the safe, economical, and efficient operation of the power system.

[0104] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0105] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical functional division. In actual implementation, there may be other division methods, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored.

[0106] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0107] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of this application.

Claims

1. A source-network collaborative planning method for the spot market, characterized in that, include: S1: Collect multi-dimensional operating parameters; S2: Process and technically transform multi-dimensional operating parameters to obtain a time-series database; S3: Construct a source-network collaborative planning model; S4: An improved particle swarm optimization algorithm is used to solve the source-network collaborative planning model with standardized datasets and K-value time series curves in the time series database as inputs, and the source-network collaborative planning scheme is obtained.

2. The source-network collaborative planning method for the spot market according to claim 1, characterized in that, In step S2, the processing and technical transformation of multi-dimensional operating parameters to obtain a time-series database includes the following steps: S21: Clean, handle missing values, and standardize the multi-dimensional operating parameters to obtain a standardized dataset; S22: Based on the supply and demand characteristics of the spot market, a supply and demand balance coefficient K is constructed as a technical indicator to obtain the time series curve of K value; S23: Store the processed standardized dataset and K-value time series curves into the time series database.

3. The source-network collaborative planning method for the spot market according to claim 2, characterized in that, In step S21, the cleaning of multi-dimensional operating parameters includes using the 3σ criterion to remove abnormal data that exceeds the equipment's rated values. Specific steps include: Select the runtime parameters to be processed, and extract the corresponding original dataset, denoted as: ； in For the first The parameter values ​​at each sampling time point, This represents the total number of samples. Based on the node type to which the parameter belongs, retrieve the rated limits of the corresponding equipment, including: extracting the maximum output limit of the equipment for the power node. , minimum output lower limit Extract the rated transmission power of the power grid nodes. Rated current Extract the maximum load that the user-side node can withstand. ; The statistical threshold intervals are calculated based on the 3σ criterion, and the intervals are as follows: ； In the above formula, The mean of the dataset; The standard deviation of the dataset; The final determination range is constructed by superimposing the equipment rating values, including: The final normal range of power node parameters ; Final normal range of grid node parameters ; The final normal range of user-side node parameters ; Abnormal data that is not within the range is removed.

4. The source-network collaborative planning method for the spot market according to claim 2, characterized in that, In step S22, the formula for calculating the K-value time series curve is as follows: ； In the formula, It represents the total real-time output of all power supply nodes in the power system; This represents the sum of real-time loads across all user-side nodes.

5. The source-network collaborative planning method for the spot market according to claim 1, characterized in that, In step S3, constructing the source-network collaborative planning model includes the following steps: S31: The objective function of the source-grid collaborative planning model is to maximize the social welfare of the power grid system, as follows: Max W ; Where W represents the social welfare of the power grid system; Annualized investment cost; Annual operating and maintenance costs; For annual penalty costs; Benefits from power output; Revenue from power grid transmission services; S32: Based on the operating rules of the power system, three types of constraints are set for the source-grid collaborative planning model, including power source constraints, grid-side constraints, and flexibility resource constraints.

6. The source-network collaborative planning method for the spot market according to claim 1, characterized in that, In step S4, the improved particle swarm optimization algorithm is used to solve the source-network collaborative planning model using a standardized dataset and K-value time series curves from the time series database as input, to obtain the source-network collaborative planning scheme. This includes the following steps: S41: Randomly generate several sets of initial planning schemes; S42: Calculate the objective function value W for each group of solutions, and select the solution with the largest W as the current optimal solution; S43: Update particle velocity and position based on the current optimal solution; S44: When the difference between the objective function values ​​of several consecutive iterations is less than the threshold, stop the iteration and output the global optimal solution; S45: Extract pure technical parameters from the global optimal solution to generate a source-network collaborative planning scheme.

7. The source-network collaborative planning method for the spot market according to claim 6, characterized in that, The formula for updating the particle velocity and position is: ； ； in, Let be the velocity of the i-th particle; ω be the inertial weight. , For learning factors; , A random number in the range [0,1]. is the individual optimal solution for the i-th particle; gbest is the global optimal solution; Let be the position of the i-th particle.

8. A source-network collaborative planning system for the spot market, characterized in that, The source-network collaborative planning method for the spot market as described in any one of claims 1 to 7 includes: The data acquisition module is used to collect multi-dimensional operating parameters. The processing module is used to process and technically transform multi-dimensional operating parameters to obtain a time-series database; The building module is used to construct the source-network collaborative planning model; The solution module is used to solve the source-network collaborative planning model by employing an improved particle swarm optimization algorithm, taking the standardized dataset and K-value time series curves in the time series database as input, and obtaining the source-network collaborative planning scheme.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the source-network collaborative planning method for the spot market as described in any one of claims 1 to 7.

10. A processor, characterized in that, The processor is used to run a program, wherein the program executes the source-network collaborative planning method for the spot market as described in any one of claims 1 to 7.

Images