New energy-thermal power collaborative optimization scheduling method based on control performance standard

Abstract

The invention discloses a new energy-thermal power collaborative optimization scheduling method based on a control performance standard, and belongs to the technical field of power system scheduling. CPS1 and CPS2 standards are converted into prospective constraints in a scheduling model in real time, and are converted into constraint conditions of a unit dynamic optimization scheduling model; the scheduling model not only considers traditional unit physical limitation and economical efficiency, but also embeds two types of core requirements of network security constraint and control performance standard to guarantee safe and stable operation of a power grid in different dimensions into an optimization decision process, so that CPS standard reaching is actively guaranteed in the scheduling process, the frequency quality of the power grid is improved, and the scheduling efficiency is improved. The dynamic characteristics of the thermal power generating unit and the second-minute level fluctuation requirements of new energy are accurately matched through an ultra-short periodic monitoring architecture combining ultra-short rolling optimization and minute-level real-time scheduling and output decision variables of the thermal power generating unit output based on a model.

Description

Technical Field

[0001] This invention relates to the field of power system dispatching technology, and more specifically, to a new energy-thermal power collaborative optimization dispatching method based on control performance standards. Background Technology

[0002] Driven by the global "dual carbon" goals and energy transition strategy, my country's new energy industry has achieved leapfrog development. As of 2024, the cumulative installed capacity of wind power and photovoltaic power in China exceeded 1.2 billion kilowatts, accounting for more than 45% of the total installed capacity. In some areas rich in new energy, the penetration rate of new energy has reached more than 60%.

[0003] However, the output of new energy sources is significantly affected by natural conditions, exhibiting strong intermittency and randomness: the minute-level power fluctuation of wind power can reach 10% to 15% of the rated capacity, and the daily output fluctuation coefficient of photovoltaic power can reach as high as 0.8. This fluctuation directly leads to the sharp rise and fall of the net load of the power grid, posing a severe challenge to the active power balance of the power grid.

[0004] To ensure the safe and stable operation of the power grid, the North American Electric Reliability Council (NERC) proposed the Control Performance Standard (CPS), which includes two core indicators, CPS1 and CPS2, to measure the quality of regulation of frequency and tie-line power within a controlled area. Most existing dispatching methods treat CPS as a post-event evaluation indicator rather than a forward-looking constraint during the dispatching process. This often leads to situations where dispatching plans exceed control deviation (ACE) limits and fail CPS assessments when facing drastic fluctuations in renewable energy sources, forcing dispatchers to make frequent and uneconomical emergency interventions.

[0005] Current research on the coordinated dispatch of new energy and thermal power mainly suffers from the following shortcomings: First, most optimization models focus on economic objectives and fail to internalize CPS requirements as hard constraints or core optimization objectives, making it difficult to proactively guarantee frequency quality in dispatch results. Second, the models do not fully consider network security constraints such as grid active power loss and line power flow, resulting in safety risks in actual operation of theoretical dispatch schemes. Finally, and most critically, existing methods have failed to effectively address the coordination and matching problem between the dynamic regulation characteristics of thermal power units (such as ramp rate and response delay) and the rapid fluctuations of new energy at the second to minute scales. This time-scale disconnect leads to a situation where regulation resources are wasted and regulation effects are poor.

[0006] To address the practical technical shortcomings, a new energy-thermal power collaborative optimization scheduling method based on control performance standards is proposed. This method aims to solve the deficiencies of traditional scheduling methods in meeting CPS standards, ensuring frequency stability, and system security after the large-scale integration of new energy into the grid. This method constructs a dynamic optimization scheduling model with control performance standard indicators as the core objective, and combines the output characteristics of thermal power and new energy to achieve refined and adaptive collaborative scheduling of the power grid. Summary of the Invention

[0007] The purpose of this invention is to address practical technical deficiencies. It provides a new energy-thermal power coordinated optimization scheduling method based on control performance standards. This method constructs a dynamic optimization scheduling model with control performance standard indicators as the core objective, dynamically coordinates the regulation characteristics of thermal power units with new energy output prediction information, and achieves the dual objectives of safe and stable grid operation and compliance with control performance standards.

[0008] The objective of this invention can be achieved through the following technical solution: a new energy-thermal power coordinated optimization scheduling method based on control performance standards, comprising the following steps: Step 1: Set an ultra-short rolling optimization cycle to obtain the ultra-short-term net load of the entire network; Step 2: Obtain the output parameters of ultra-short-term thermal power units within the same timestamp; Step 3: Dynamically adjust the output decision variables of each thermal power unit based on control performance constraints: Construct and solve a dynamic optimization scheduling model of the units with the goal of minimizing the total operating cost and including control performance constraint indicators as constraints. Set the basic scheduling period at the minute level, and take the absolute value of the net load of the entire network in the ultra-short term, the output parameters of the thermal power units in the ultra-short term, the control performance constraint indicators obtained based on the control performance standard conversion, and the minimum target operating cost as input features. The model outputs the output decision variables of each thermal power unit in the future multiple basic scheduling periods. Step 4: Scheduling Execution: Based on the output decision variables of each thermal power unit during the first basic scheduling period, scheduling instructions are generated and issued to each thermal power unit for execution to adjust its operating status. The execution status is continuously monitored in preparation for the next cycle of optimization.

[0009] Furthermore, in step one, the predicted grid load, the predicted ultra-short-term output of renewable energy power plants, the planned power of tie lines, and the active power loss of the grid within the ultra-short rolling optimization period are obtained. The ultra-short-term net load of the entire grid is calculated using the formula: predicted grid load + active power loss of the grid - predicted ultra-short-term output of renewable energy power plants - planned power of tie lines.

[0010] Furthermore, in step two, the output parameters of ultra-short-term thermal power units include the initial output value of each thermal power unit, the upper and lower limits of the output of each thermal power unit, the ramp rate of each thermal power unit, the maximum constrained ramp rate of each thermal power unit, and the logical constraints of the ramp direction indicator variable.

[0011] Furthermore, in step three, a dynamic optimization scheduling model for the generating units is constructed using control performance constraints and multiple safety constraints. The control performance constraints are implemented by converting control performance standards CPS1 and CPS2 into time-varying limits for regional control deviation ACE. The multiple safety constraints are equality constraints on system power balance and dynamic changes in tie line power, as well as inequality constraints on thermal power unit output and ramp rate, minimum continuous adjustment time, branch and cross-sectional power flow safety, and tie line power deviation.

[0012] Furthermore, the system frequency deviation is the difference between the actual frequency of the power grid and the rated frequency, and the area control deviation is the difference between the actual power of the tie line and the planned power of the tie line plus the product of the system frequency deviation and the frequency deviation coefficient, where the frequency deviation coefficient is a constant.

[0013] Furthermore, the output decision variables include the output that each unit needs to adjust relative to the initial output value, positive ramp direction indicator variables, negative ramp direction indicator variables, and ramp rate allocation variables.

[0014] Furthermore, the ultra-short rolling optimization cycle is 15 minutes, and the basic scheduling period is 1 minute.

[0015] Compared with the prior art, the advantages of this invention are: This invention transforms the CPS1 and CPS2 standards into forward-looking constraints (i.e., ACE time-varying limit constraints) in the scheduling model in real time, and into constraints of the dynamic optimization scheduling model of the units. This scheduling model not only considers the traditional physical limitations and economics of the units, but more importantly, it embeds the two core requirements for ensuring the safe and stable operation of the power grid in different dimensions, namely "network security constraints" and "control performance standards", into the optimization decision-making process, so as to proactively ensure that CPS standards are met during the scheduling process and improve the frequency quality of the power grid.

[0016] This invention also employs an ultra-short-cycle monitoring architecture that combines "15-minute ultra-short rolling optimization" with "1-minute real-time scheduling," and introduces key output decision variables to accurately match the dynamic characteristics of thermal power units with the second-minute fluctuations in demand from new energy sources. This avoids ineffective and reverse adjustment actions of thermal power units, thereby solving the problem of dynamic resource coordination across multiple time scales and significantly improving the economic efficiency of regulation. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0018] 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 embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0019] Example 1: This invention discloses a new energy-thermal power coordinated optimization scheduling method based on control performance standards. Please refer to [link / reference]. Figure 1 This includes the following steps: Step 1: Set an ultra-short rolling optimization cycle to obtain the ultra-short-term net load of the entire network; Step 2: Obtain the output parameters of ultra-short-term thermal power units within the same timestamp; Step 3: Dynamically adjust the output decision variables of each thermal power unit based on control performance constraints: Construct and solve a dynamic optimization scheduling model of the units with the goal of minimizing the total operating cost and including control performance constraint indicators as constraints. Set the basic scheduling period at the minute level, and take the absolute value of the net load of the entire network in the ultra-short term, the output parameters of the thermal power units in the ultra-short term, the control performance constraint indicators obtained based on the control performance standard conversion, and the minimum target operating cost as input features. The model outputs the output decision variables of each thermal power unit in the future multiple basic scheduling periods. Step 4: Scheduling Execution: Based on the output decision variables of each thermal power unit during the first basic scheduling period, scheduling instructions are generated and issued to each thermal power unit for execution to adjust its operating status. The execution status is continuously monitored in preparation for the next cycle of optimization.

[0020] After the scheduling command is executed, the system state changes. The new system frequency deviation is obtained through real-time measurement and the new regional control deviation (ACE) is calculated. This information is fed back to the model at the beginning of the next scheduling period, triggering a new round of rolling optimization. This is the core of the method's ability to adapt to new energy fluctuations and maintain high performance.

[0021] In step one, the predicted grid load, the predicted ultra-short-term output of renewable energy power plants, the planned power of tie lines, and the active power loss of the grid within the ultra-short rolling optimization period are obtained. The ultra-short-term net load of the entire grid is calculated using the formula: predicted grid load + grid active power loss - predicted ultra-short-term output of renewable energy power plants - planned power of tie lines. Here, we assume that the net load is positive, so the output of thermal power units needs to be increased. If the net load is negative, the output needs to be reduced.

[0022] In step two, the output parameters of ultra-short-term thermal power units include the initial output value of each thermal power unit, the upper and lower limits of the output of each thermal power unit, the ramp rate of each thermal power unit, the maximum constraint ramp rate of each thermal power unit, and the logical constraints of the ramp direction indicator variable.

[0023] In step three, a dynamic optimization scheduling model for generating units is constructed using control performance constraints and multiple safety constraints. The control performance constraints are implemented by converting control performance standards CPS1 and CPS2 into time-varying limits for regional control deviation ACE. The system frequency deviation is the difference between the actual frequency of the power grid and the rated frequency. The regional control deviation is the difference between the actual power and the planned power of the tie line, plus the product of the system frequency deviation and the frequency deviation coefficient. The frequency deviation coefficient is a constant approved and set in the control system by the power grid dispatching agency. These indicators are reflected in real time during the operation of the power system and are measured and recorded by the monitoring system. The multiple safety constraints are the equality constraints of system power balance and dynamic changes in tie line power, as well as the inequality constraints of thermal power unit output and ramp rate, minimum continuous adjustment time, branch and cross-sectional power flow safety, and tie line power deviation. By converting the CPS1 and CPS2 standards into forward-looking constraints (i.e., ACE time-varying limit constraints) in the scheduling model in real time, scheduling instructions are proactively compatible with frequency stability requirements during the generation phase. This transforms "post-event remediation" into "pre-event prevention," fundamentally improving the frequency regulation quality and CPS compliance rate of the power grid, enhancing the grid's ability to withstand power fluctuations from new energy sources, and incorporating multiple security constraints. This ensures that the generated scheduling plan not only adapts to the uncertainty of new energy output but also proactively avoids operational risks such as line overload, guaranteeing the real-time operational safety of the power grid under conditions of strong fluctuations in new energy power generation.

[0024] In step four, the output decision variables include the output that each unit needs to adjust relative to the initial output value, the positive ramp direction indicator variable, the negative ramp direction indicator variable, and the ramp rate allocation variable. The ramp direction indicator variable can be obtained from the difference between the adjusted output and the current output, but an indicator variable can also be output, for example, 1 indicates upward adjustment and 0 indicates downward adjustment. Based on model-based optimization scheduling, we seek a scheduling scheme that can satisfy power balance (tracking net load), without violating any safety constraints, and at the same time optimize CPS indicators. This is equivalent to a continuous and automatic "comparison and fusion" process. The essence of the whole optimization scheduling is to allow adjustable thermal power to match the fluctuating "net load" determined by both load and new energy sources, and to find the optimal matching method under the premise of meeting CPS standards and network security.

[0025] Furthermore, it should be noted that the ultra-short rolling optimization cycle under multiple safety constraints is 15 minutes, and the basic scheduling period under multiple safety constraints is 1. The ultra-short rolling optimization cycle is mainly a "forward-looking decision window," the significance of which is to match the reliability of predictions: the ultra-short-term predictions of new energy (wind and solar) (such as 0-4 hour predictions) have relatively high accuracy on a 15-minute scale. Using this as an optimization window, reliable prediction information can be effectively used to formulate scheduling strategies. Within the ultra-short rolling optimization cycle of the same timestamp, the unit regulation characteristics are compatible. Physical constraints such as the ramp rate and minimum start-stop time of thermal power units will only have a significant impact on a time scale of several minutes to tens of minutes. The 15-minute window is sufficient to capture and optimize these dynamic processes. To achieve the "rolling optimization" strategy, the scheduling model does not calculate only once every 15 minutes, but every minute (i.e., one basic scheduling period), it re-optimizes the next 15 minutes starting from the current moment. In this way, decisions can always be made based on the latest system status (measured frequency, ACE, actual unit output, etc.), forming closed-loop control and enhancing anti-interference capabilities.

[0026] A basic scheduling period of 1 minute is set to align with the assessment cycle of the CPS standard, enabling rapid and accurate minute-level response to system frequency and regional control deviation (ACE). The optimization model is executed once per minute based on the latest system state, forming an automatic control process of "forward decision-making - closed-loop correction," thereby ensuring high-level compliance with the CPS standard while coping with the strong volatility of new energy sources.

[0027] In summary, this invention transforms the CPS1 and CPS2 standards into forward-looking constraints (i.e., ACE time-varying limit constraints) in the scheduling model in real time, and into constraints for the dynamic optimization scheduling model of the units. This scheduling model not only considers the traditional physical limitations and economics of the units, but more importantly, it embeds the two core requirements for ensuring the safe and stable operation of the power grid in different dimensions, namely "network security constraints" and "control performance standards", into the optimization decision-making process. This enables proactively ensuring CPS compliance during the scheduling process and improves the frequency quality of the power grid. By combining "15-minute ultra-short rolling optimization" with "1-minute real-time scheduling" into an ultra-short cycle monitoring architecture, and by introducing multiple key output decision variables, the dynamic characteristics (climbing rate, response time) of thermal power units are accurately matched with the second-minute fluctuations in the demand of new energy sources. This avoids ineffective and reverse adjustment actions of thermal power units, solves the problem of dynamic resource coordination under multiple time scales, and significantly improves the economic efficiency of regulation.

[0028] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto; any equivalent substitutions or modifications made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and its improved concept, should be covered within the scope of protection of the present invention.

Claims

1. A new energy-thermal power coordinated optimization scheduling method based on control performance standards, characterized in that: Includes the following steps: Step 1: Set an ultra-short rolling optimization cycle to obtain the ultra-short-term net load of the entire network; Step 2: Obtain the output parameters of ultra-short-term thermal power units within the same timestamp; Step 3: Dynamically adjust the output decision variables of each thermal power unit based on control performance constraints: Construct and solve a dynamic optimization scheduling model of the units with the goal of minimizing the total operating cost and including control performance constraint indicators as constraints. Set the basic scheduling period at the minute level, and take the absolute value of the net load of the entire network in the ultra-short term, the output parameters of the thermal power units in the ultra-short term, the control performance constraint indicators obtained based on the control performance standard conversion, and the minimum target operating cost as input features. The model outputs the output decision variables of each thermal power unit in the future multiple basic scheduling periods. Step 4: Scheduling Execution: Based on the output decision variables of each thermal power unit during the first basic scheduling period, scheduling instructions are generated and issued to each thermal power unit for execution to adjust its operating status. The execution status is continuously monitored in preparation for the next cycle of optimization.

2. The new energy-thermal power coordinated optimization scheduling method based on control performance standards according to claim 1, characterized in that: In step one, the predicted grid load, the predicted ultra-short-term output of renewable energy power plants, the planned power of tie lines, and the active power loss of the grid within the ultra-short rolling optimization period are obtained. The ultra-short-term net load of the entire grid is calculated using the formula: predicted grid load + grid active power loss - predicted ultra-short-term output of renewable energy power plants - planned power of tie lines.

3. The new energy-thermal power coordinated optimization scheduling method based on control performance standards according to claim 1, characterized in that: In step two, the output parameters of ultra-short-term thermal power units include the initial output value of each thermal power unit, the upper and lower limits of the output of each thermal power unit, the ramp rate of each thermal power unit, the maximum constraint ramp rate of each thermal power unit, and the logical constraints of the ramp direction indicator variable.

4. The new energy-thermal power coordinated optimization scheduling method based on control performance standards according to claim 1, characterized in that: In step three, a dynamic optimization scheduling model for the generating units is constructed using control performance constraints and multiple safety constraints. The control performance constraints are implemented by converting the control performance standards CPS1 and CPS2 into time-varying limits for the regional control deviation ACE. The multiple safety constraints are equality constraints on system power balance and dynamic changes in tie line power, as well as inequality constraints on the output and ramp rate of thermal power units, minimum continuous adjustment time, power flow safety of branches and sections, and power deviation from tie lines.

5. The new energy-thermal power coordinated optimization scheduling method based on control performance standards according to claim 4, characterized in that: The system frequency deviation is the difference between the actual frequency of the power grid and the rated frequency. The area control deviation is the difference between the actual power of the tie line and the planned power of the tie line, plus the product of the system frequency deviation and the frequency deviation coefficient, where the frequency deviation coefficient is a constant.

6. The new energy-thermal power coordinated optimization scheduling method based on control performance standards according to claim 1, characterized in that: The output decision variables include the output that each unit needs to adjust relative to the initial output value, positive ramp direction indicator variables, negative ramp direction indicator variables, and ramp rate allocation variables.

7. The new energy-thermal power coordinated optimization scheduling method based on control performance standards according to claim 1, characterized in that: The ultra-short rolling optimization cycle is 15 minutes, and the basic scheduling period is 1 minute.

Images