Multi-time-scale coordinated optimization scheduling method for power distribution network

Abstract

The invention relates to the technical field of power distribution network multi-time scale scheduling, in particular to a power distribution network multi-time scale coordinated optimization scheduling method. According to the technical scheme, the method comprises long-time-scale optimization scheduling, short-time-scale optimization scheduling and a coordination mechanism, and the long-time-scale optimization scheduling and the short-time-scale optimization scheduling are matched to form the power distribution network multi-time-scale coordination optimization scheduling method. And the coordination mechanism performs seamless connection and overall optimization on the long-time scale optimization scheduling and the short-time scale optimization scheduling. The multi-time-scale coordinated optimization scheduling method for the power distribution network has the advantages of improving system operation efficiency, enhancing system stability, improving economy and flexibility, comprehensively considering multi-objective optimization and the like. Due to the advantages, the method has important application value and development prospect in optimal scheduling of a modern electric power system.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-time scale dispatching of distribution networks, and in particular to a multi-time scale coordinated optimization dispatching method for distribution networks. Background Art

[0002] The distribution network is a key component of the power system, responsible for transmitting and distributing the electricity generated by power plants to various users and devices. It ensures that electricity is stably and efficiently transmitted to every corner that needs electricity through cables, wires, transformers and other equipment. Through reasonable planning and construction of the distribution network, it can ensure the effective access and efficient use of new energy generation and promote the widespread application of clean energy.

[0003] The traditional grid dispatching method increases the uncertainty of the grid when facing the access of renewable energy (such as wind power and photovoltaic power). At the same time, the response speed is slow when the grid fails and the control level of load in the grid is low.

[0004] After searching, the publication number CN104993522B discloses a multi-time scale coordinated optimization scheduling method for an active distribution network based on MPC, which includes the following steps: (1) performing long-time scale optimization scheduling on the active distribution network; (2) performing short-time scale rolling optimization scheduling on the active distribution network based on MPC as a comparison with the present invention according to the optimization result of the long-time scale optimization scheduling of the active distribution network.

[0005] To this end, we propose a multi-time scale coordinated optimization scheduling method for distribution networks to solve the existing problems. Summary of the invention

[0006] The purpose of the present invention is to propose a multi-time scale coordinated optimization scheduling method for distribution network in view of the problems existing in the background technology.

[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a distribution network multi-time scale coordinated optimization scheduling method, including long time scale optimization scheduling, short time scale optimization scheduling and a coordination mechanism, wherein the long time scale optimization scheduling is coordinated with the short time scale optimization scheduling to constitute a distribution network multi-time scale coordinated optimization scheduling method, and the coordination mechanism performs seamless connection and overall optimization of the long time scale optimization scheduling and the short time scale optimization scheduling.

[0008] Preferably, the long-time scale optimization scheduling includes data collection and prediction, model building and algorithm solving, and the data collection and prediction collects data such as the architecture parameters of the distribution network, load prediction, and distributed energy prediction, wherein the architecture parameters include the resistance, reactance, length, and cross-sectional area of ​​the distribution line, the transformation ratio, loss, and capacity of the transformer, the load size, type (industrial, commercial, residential), and power factor of the load node, and the power generation capacity and power generation type (thermal power, wind power, photovoltaic, etc.) of the power generation node;

[0009] Load forecasting includes time scales of hours and minutes, different geographical regions, different load nodes, seasonal changes of spring, summer, autumn and winter, working days and weekends, past load data from historical data analysis, and the impact of holidays, etc.

[0010] Distributed energy forecasts include solar energy intensity, solar panel area, conversion efficiency, etc., wind energy wind speed, wind turbine capacity, wind turbine conversion efficiency, etc., weather factors such as temperature and humidity, equipment status failure rate, maintenance plan, power generation attenuation, etc.

[0011] Preferably, the model is established based on the collected data to establish a long-time scale optimization scheduling model, including objective functions and constraints, and a multi-time scale distribution scheduling model (Multi-time Scale Distribution Scheduling Model) is adopted, whose objective functions include minimizing operating costs, minimizing carbon emissions, maximizing renewable energy utilization, and minimizing grid voltage drop, and the constraints include power balance constraints, power generation limitation constraints, load limitation constraints, line capacity constraints, node voltage constraints, switch state constraints, battery state constraints, and operation constraints.

[0012] Preferably, the algorithm solves the problem using a linear programming algorithm, and the specific steps of the linear programming algorithm include defining the problem, standardizing the form, constructing an initial simplex table, selecting basis entry and exit variables, updating the simplex table, checking optimality, and iterating the process, reading the optimal solution (the value of the decision variable) and the optimal target value from the final simplex table, and analyzing the impact of changes in the objective function coefficients and constraints on the optimal solution.

[0013] Preferably, the short-time scale optimization scheduling includes real-time data update, prediction model adjustment, rolling optimization and feedback and correction, wherein the real-time data update must ensure timely collection of various real-time data, such as equipment status, production progress, energy consumption, etc., use sensors, monitoring equipment and other technical means to acquire data in real time, ensure that real-time data can be quickly transmitted to the scheduling system, and use efficient communication technologies, such as cloud computing, the Internet of Things, etc., to ensure the real-time and stability of data transmission.

[0014] Preferably, the forecasting model is a short-term load forecasting model (Short-Term Load Forecasting Model), whose objective functions include cost minimization, grid security and renewable energy utilization, and whose constraints include power balance constraints, generator constraints, transmission line constraints, and load constraints. By modeling and optimizing the objective function and constraints, accurate and effective scheduling of short-term load forecasting can be achieved, ensuring the safe and stable operation of the power system and minimizing operating costs.

[0015] Preferably, the rolling optimization adopts a linear programming algorithm, and the specific steps of the linear programming algorithm include defining the problem, standardizing the form, constructing an initial simplex table, selecting entry and exit variables, updating the simplex table, checking optimality, and iterating the process. The optimal solution (the value of the decision variable) and the optimal target value are read from the final simplex table, and the impact of changes in the objective function coefficients and constraint conditions on the optimal solution are analyzed. Short-time-scale rolling optimization is performed to generate scheduling instructions at the current moment. The feedback and correction are used to correct the prediction model and the rolling optimization strategy according to the execution results and feedback information.

[0016] Preferably, the coordination mechanism includes information sharing and transmission, coordination of dispatching strategies, technical support and collaborative work. Through the above mechanism, information sharing, dispatching strategy coordination and technical support can be achieved between long-term dispatching and short-term dispatching, thereby ensuring the safe, economical and reliable operation of the power system.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The present invention discloses a multi-time scale coordinated optimization scheduling method for a distribution network. The method can more accurately match the power supply and load requirements by optimizing scheduling at different time scales, reduce energy waste, and thus improve the overall efficiency of system operation. An ultra-short period rolling optimization strategy is adopted in electric energy scheduling, and the scheduling plan can be adjusted in real time to cope with the rapid changes in load and power generation power, thereby ensuring the reliability and economy of power supply. Multi-time scale scheduling can better match source-load characteristics at different time scales, and reduce energy waste and cost increase caused by mismatch between supply and demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the flow chart structure of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Embodiment 1

[0022] like Figure 1 As shown, the present invention proposes a distribution network multi-time scale coordinated optimization scheduling method, including long time scale optimization scheduling, short time scale optimization scheduling and coordination mechanism. The long time scale optimization scheduling and the short time scale optimization scheduling are coordinated to form a distribution network multi-time scale coordinated optimization scheduling method. The coordination mechanism seamlessly connects and comprehensively optimizes the long time scale optimization scheduling and the short time scale optimization scheduling.

[0023] Based on Example 1:

[0024] Long-term optimal scheduling is day-ahead scheduling, which is based on the load forecast and power generation capacity forecast for the next 24 hours or longer to formulate a preliminary power generation plan and power grid scheduling plan. Short-term optimal scheduling is intraday scheduling, which is based on day-ahead scheduling and more accurate short-term load forecasts and power generation capacity forecasts to make rolling corrections and adjustments to power generation plans and power grid scheduling plans. In the operation and management of distribution networks, the distribution network is coordinated and optimized by combining the needs and constraints of long and short time scales to achieve the goals of energy supply and demand balance, safe and stable operation, and economical and efficient operation. This model can comprehensively consider factors such as the power market, energy management, and equipment operation to improve the operating efficiency and reliability of the distribution network.

[0025] Embodiment 2

[0026] like Figure 1 As shown, a multi-time scale coordinated optimization scheduling method for a distribution network proposed by the present invention, compared with the first embodiment, this embodiment also includes: long-time scale optimization scheduling includes data collection and prediction, model establishment and algorithm solution, data collection and prediction collect data such as architecture parameters, load prediction, and distributed energy prediction of the distribution network, wherein the architecture parameters include resistance, reactance, length, cross-sectional area, etc. of the distribution line, transformation ratio, loss, capacity, etc. of the transformer, load size, type (industrial, commercial, residential), power factor, etc. of the load node, power generation capacity, power generation type (thermal power, wind power, photovoltaic, etc.) of the power generation node;

[0027] Load forecasting includes time scales of hours and minutes, different geographical regions, different load nodes, seasonal changes of spring, summer, autumn and winter, working days and weekends, past load data from historical data analysis, and the impact of holidays, etc.

[0028] Distributed energy forecasting includes solar energy intensity, solar panel area, conversion efficiency, wind speed, wind turbine capacity, wind turbine conversion efficiency, temperature, humidity, equipment failure rate, maintenance plan, power generation attenuation, etc. The model is established based on the collected data to establish a long-term optimization scheduling model, including objective functions and constraints, and adopts a multi-time scale distribution scheduling model (Multi-time Scale Distribution Scheduling Model), whose objective function includes minimizing operating costs including power generation cost, power purchase cost, line loss cost, etc., minimizing carbon emissions taking into account environmental factors, reducing carbon emissions, maximizing renewable energy utilization and optimizing scheduling, increasing renewable energy consumption, and minimizing grid voltage drop to reduce grid voltage drop and improve power supply quality. Constraints include power balance constraints to ensure supply and demand balance; power generation restriction constraints taking into account generator capacity, minimum operating restrictions, etc.; load restriction constraints taking into account load demand, power factor, etc.; line capacity constraints taking into account line heat loss, voltage stability, etc.; node voltage constraints ensuring that node voltage is within a reasonable range; switch state constraints taking into account switch state, operation constraints, etc.; battery state constraints taking into account battery charge and discharge state, etc. and operation constraints taking into account equipment start and stop, switching and other operation constraints;

[0029] The function formula is shown as follows:

[0030]

[0031] Where T is a time set, which may include multiple time periods before and during the day;

[0032] C gen (t) is the electricity generation cost during period t;

[0033] C loss (t) is the network loss cost during period t;

[0034] C storage (t) is the operating cost of the energy storage equipment during period t;

[0035] …indicates other possible cost items.

[0036] Power balance constraints:

[0037] P gen (t)+P storage，dis (t)-P storage，chg (t)-P load (t) = 0

[0038] P gen (t) is the power generated during period t

[0039] P storage，dis (t) and P storage，chg (t) are the energy storage discharge and charging power during period t

[0040] P load (t) is the load power during period t;

[0041] Minimize the grid voltage drop and reduce the grid voltage drop. Objective function 2: The deviation between the voltage and the rated voltage is minimized as the optimization target as shown in the formula:

[0042]

[0043] Where V i (t) is the voltage amplitude of the ith node at time t, and nbus is the number of nodes.

[0044] Embodiment 3

[0045] Short-term optimization scheduling includes real-time data update, forecast model adjustment, rolling optimization, feedback and correction. Real-time data update should ensure timely collection of various real-time data, such as equipment status, production progress, energy consumption, etc., use sensors, monitoring equipment and other technical means to obtain data in real time, ensure that real-time data can be quickly transmitted to the scheduling system, and use efficient communication technologies, such as cloud computing and the Internet of Things, to ensure the real-time and stability of data transmission;

[0046] The forecasting model is a short-term load forecasting model. Its objective functions include cost minimization, grid security and renewable energy utilization. Its constraints include power balance constraints, generator constraints, transmission line constraints and load constraints. By modeling and optimizing the objective functions and constraints, it is possible to achieve accurate and effective scheduling of short-term load forecasting, ensure the safe and stable operation of the power system, and minimize operating costs.

[0047] Its function formula is:

[0048]

[0049] in is the predicted load value at time t;

[0050] β 0 , β 1 , ..., β n are the parameters of the model, which need to be estimated through training data;

[0051] x 1，t , x 2，t , ..., x n，t It is the input feature at time t, which may include historical load data, weather conditions (such as temperature, humidity, wind speed, etc.), time characteristics (such as hours, days of the week, months, holidays, etc.), economic indicators, etc. It aims to predict the power load demand in the next few hours to days. Since power load is affected by many factors, such as weather, holidays, economic activities, time patterns, etc., short-term load forecasting models usually need to consider these factors.

[0052] The algorithm solution of the long-time scale optimization scheduling model and the rolling optimization of the short-time scale optimization scheduling both adopt the linear programming algorithm. The specific steps of the linear programming algorithm include defining the problem, standardizing the form, constructing the initial simplex table, selecting the entry and exit variables, updating the simplex table, checking the optimality, iterative process, reading the optimal solution (the value of the decision variable) and the optimal target value from the final simplex table, analyzing the impact of the objective function coefficient and the change of the constraint conditions on the optimal solution, performing the rolling optimization of the short-time scale, generating the scheduling instructions at the current moment, feedback and correction. According to the execution results and feedback information, the prediction model and the rolling optimization strategy are corrected;

[0053] The linear programming algorithm can be expressed as an interior point algorithm:

[0054]

[0055] stAx=b

[0056] x≥0

[0057] where c∈R n , A∈R m×n , b∈R m , and x is the decision variable. A logarithmic barrier term is introduced to penalize any element in the solution vector x that is close to zero, thereby keeping the solution inside the feasible region.

[0058] Embodiment 4

[0059] The coordination mechanism includes information sharing and transmission, coordination of dispatching strategies, technical support and collaborative work. Through the above mechanisms, information sharing, dispatching strategy coordination and technical support can be achieved between long-term dispatching and short-term dispatching, so as to ensure the safe, economical and reliable operation of the power system. It is necessary to establish a unified information platform to integrate data of various time scales, including load forecasting, power generation capacity, energy storage status, market prices, etc., to achieve real-time information sharing, data preprocessing and fusion: clean, verify and fuse data from different sources and with different precision to improve data accuracy and reliability;

[0060] Goal alignment: Ensure that the optimization goals at each time scale (such as minimizing costs, maximizing reliability, reducing carbon emissions, etc.) are consistent with the overall goal.

[0061] Constraint transfer: gradually refine the long-term planning constraints (such as equipment capacity limitations, environmental protection requirements, etc.) and transfer them to short-term scheduling and real-time control to ensure that decisions at all time scales meet these constraints;

[0062] Distributed coordination: For large-scale distribution networks, distributed optimization algorithms can be used to decompose the large system into multiple subsystems for parallel computing to improve computing efficiency.

[0063] Centralized coordination: Global optimization is performed at key nodes or control centers to ensure coordination among subsystems and avoid local optimal solutions. It is necessary to comprehensively consider the characteristics and needs of different time scales, and achieve safe, economical and efficient operation of the power grid through information interaction, target coordination, rolling optimization, distributed and centralized coordination, and emergency response.

[0064] The above-mentioned specific embodiments are only several preferred embodiments of the present invention. Based on the technical solutions of the present invention and the relevant inspirations of the above-mentioned embodiments, those skilled in the art can make various alternative improvements and combinations to the above-mentioned specific embodiments.

[0065] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A multi-time scale coordinated optimization dispatching method for a distribution network, comprising long-time scale optimization dispatching, short-time scale optimization dispatching and a coordination mechanism, characterized in that: The long-time scale optimization scheduling is coordinated with the short-time scale optimization scheduling to form a multi-time scale coordinated optimization scheduling method for the distribution network. The coordination mechanism seamlessly connects and optimizes the long-time scale optimization scheduling and the short-time scale optimization scheduling as a whole, thereby achieving the purpose of improving system operation efficiency, enhancing system stability, promoting new energy consumption, and meeting diversified needs.

2. A distribution network multi-time scale coordinated optimization scheduling method according to claim 1, characterized in that: The long-term optimal scheduling includes data collection and prediction, model building and algorithm solving. The data collection and prediction collects data such as the architecture parameters of the distribution network, load prediction, and distributed energy prediction. The architecture parameters include the resistance, reactance, length, and cross-sectional area of ​​the distribution line, the transformation ratio, loss, and capacity of the transformer, the load size, type (industrial, commercial, residential), and power factor of the load node, and the power generation capacity and power generation type (thermal power, wind power, photovoltaic, etc.) of the power generation node. Load forecasting includes time scales of hours and minutes, different geographical regions, different load nodes, seasonal changes of spring, summer, autumn and winter, working days and weekends, past load data from historical data analysis, and the impact of holidays, etc. Distributed energy forecasts include solar energy intensity, solar panel area, conversion efficiency, etc., wind energy wind speed, wind turbine capacity, wind turbine conversion efficiency, etc., weather factors such as temperature and humidity, equipment status failure rate, maintenance plan, power generation attenuation, etc.

3. A distribution network multi-time scale coordinated optimization scheduling method according to claim 2, characterized in that: The model is established based on the collected data to establish a long-time scale optimization scheduling model, including objective functions and constraints, and adopts a multi-time scale distribution scheduling model (Multi-time Scale Distribution Scheduling Model). Its objective functions include minimizing operating costs, minimizing carbon emissions, maximizing renewable energy utilization, and minimizing grid voltage drop. The constraints include power balance constraints, power generation limitation constraints, load limitation constraints, line capacity constraints, node voltage constraints, switch state constraints, battery state constraints, and operation constraints.

4. A distribution network multi-time scale coordinated optimization scheduling method according to claim 2, characterized in that: The algorithm adopts a linear programming algorithm for solving the problem. The specific steps of the linear programming algorithm include defining the problem, standardizing the form, constructing an initial simplex table, selecting entry and exit variables, updating the simplex table, checking optimality, and iterating the process. The optimal solution (the value of the decision variable) and the optimal target value are read from the final simplex table, and the influence of changes in the objective function coefficients and constraint conditions on the optimal solution is analyzed.

5. A distribution network multi-time scale coordinated optimization scheduling method according to claim 1, characterized in that: The short-time scale optimization scheduling includes real-time data update, prediction model adjustment, rolling optimization and feedback and correction. The real-time data update must ensure the timely collection of various real-time data, such as equipment status, production progress, energy consumption, etc., use sensors, monitoring equipment and other technical means to obtain data in real time, ensure that real-time data can be quickly transmitted to the scheduling system, and use efficient communication technologies, such as cloud computing and the Internet of Things, to ensure the real-time and stability of data transmission.

6. A distribution network multi-time scale coordinated optimization scheduling method according to claim 5, characterized in that: The forecasting model is a short-term load forecasting model (Short-Term Load Forecasting Model), whose objective functions include cost minimization, grid security and renewable energy utilization, and whose constraints include power balance constraints, generator constraints, transmission line constraints, and load constraints. By modeling and optimizing the objective function and constraints, accurate and effective scheduling of short-term load forecasting can be achieved, ensuring the safe and stable operation of the power system and minimizing operating costs.

7. A distribution network multi-time scale coordinated optimization scheduling method according to claim 5, characterized in that: The rolling optimization adopts a linear programming algorithm. The specific steps of the linear programming algorithm include defining the problem, standardizing the form, constructing an initial simplex table, selecting entry and exit variables, updating the simplex table, checking optimality, and iterating the process. The optimal solution (the value of the decision variable) and the optimal target value are read from the final simplex table, the impact of changes in the objective function coefficients and constraint conditions on the optimal solution are analyzed, and rolling optimization is performed on a short time scale to generate a scheduling instruction at the current moment. The feedback and correction are used to correct the prediction model and the rolling optimization strategy based on the execution results and feedback information.

8. A distribution network multi-time scale coordinated optimization scheduling method according to claim 1, characterized in that: The coordination mechanism includes information sharing and transmission, coordination of dispatching strategies, technical support and collaborative work. Through the above mechanism, information sharing, dispatching strategy coordination and technical support can be achieved between long-term dispatching and short-term dispatching, thereby ensuring the safe, economical and reliable operation of the power system.

Images