Power distribution network planning method based on Monte Carlo model

Abstract

The invention provides a power distribution network planning method based on a Monte Carlo model, and belongs to the technical field of power distribution network planning. In order to solve the technical problem that a decision maker cannot make a more accurate power distribution network planning scheme on the basis of potential benefits brought by various technical development paths at present, the adopted technical scheme is as follows: defining targets and constraint conditions of power distribution network planning as follows: based on constraint conditions of voltage deviation, line capacity, short-circuit current and an N-1 criterion, determining a power distribution network planning model; the planning targets of minimizing the investment cost, maximizing the reliability and optimizing the renewable energy source permeability are achieved; establishing a probability model for the key uncertainty factors by adopting a Monte Carlo model based on probability statistics and stochastic simulation, generating various possible load and renewable energy output scenes in the future through the Monte Carlo model, and generating a random scene combined by multiple groups of random parameters through probability distribution; the method is applied to power distribution network planning.

Description

Technical Field

[0001] The present invention provides a distribution network planning method based on a Monte Carlo model, belonging to the technical field of distribution network planning. Background Art

[0002] At present, with the continuous advancement of distribution network construction, higher requirements have been put forward for distribution network construction planning. Investment planning has shifted from coarse to fine, from quantity to quality, and from pursuing effects to pursuing benefits. The transformation and investment decisions of distribution networks will also follow the construction principles of safety, reliability, economy and efficiency, advanced technology and environmental friendliness. In addition, with the access and operation of new distribution networks such as renewable energy access and the popularization of electric vehicles in recent years, it is necessary to reveal the potential risks of planning schemes through probabilistic analysis to ensure the economy and stability of distribution network planning and construction. However, the decision makers currently rely on cannot make more accurate planning schemes based on the potential benefits brought by various technological development paths, and cannot meet the expected indicators of distribution network performance. Summary of the Invention

[0003] In order to solve the technical problems existing in the background technology, the present invention adopts a technical solution: providing a distribution network planning method based on a Monte Carlo model, comprising the following planning steps: Step 1: Define the distribution network planning objectives and constraints: Based on the constraints of voltage deviation, line capacity, short-circuit current, and the N-1 criterion, achieve the planning goals of minimizing investment costs, maximizing reliability, and optimizing renewable energy penetration; In distribution network planning, a Monte Carlo model based on probability statistics and stochastic simulation is used to establish a probability model for the following key uncertainties, including: Establish load demand models based on historical data or forecast models; Establish a distributed energy model based on the output fluctuations of photovoltaic and wind power; Establish equipment failure models based on line and transformer failure rates; Establish electricity price and policy models based on market electricity price fluctuations and subsidy policy changes; Step 2: Generate multiple possible future load and renewable energy output scenarios through the Monte Carlo model, and generate multiple sets of random parameter combinations through probability distribution; Step 3: Perform the following processing on each random scene: Power flow calculation: verify whether voltage and power exceed the limit; Reliability assessment: Calculate power outage frequency and duration; Economic analysis: Calculate investment costs, operation and maintenance costs, and renewable energy benefits; Step 4: Calculate and count the mean, variance, and quantile of the objective function, draw a distribution histogram or calculate the confidence interval to identify high-risk scenarios; The planning scheme with the best robustness is selected based on statistical results, or the optimal planning scheme is calculated by combining it with a random optimization algorithm.

[0004] The power flow calculation performed in step 3 specifically includes: performing power flow calculation for each scenario, evaluating whether constraints such as voltage and line capacity are met, and calculating reliability and economic indicators of different planning schemes.

[0005] In step 3, the power outage frequency and power outage time are calculated by using a Monte Carlo model to perform fault simulation.

[0006] The random optimization algorithm used in step 4 is specifically a random programming algorithm or a chance constrained programming algorithm.

[0007] The beneficial effects of the present invention compared with the existing technology are as follows: the present invention adopts the Monte Carlo model to provide quantitative uncertainty for distribution network planning, which is particularly suitable for distribution network planning in new distribution network scenarios such as high proportion of renewable energy access and popularization of electric vehicles. Its core value lies in revealing the potential risks of planning schemes through probability analysis, and assisting decision makers to achieve the optimal trade-off between economy and reliability; the present invention adopts an active distribution network planning method based on Monte Carlo tree search and reinforcement learning algorithm, and deduces the potential benefits brought by different technology development paths through Monte Carlo tree search and reinforcement learning algorithm, helping decision makers to make more accurate planning schemes, so that the distribution network performance can meet the expected indicators. DETAILED DESCRIPTION

[0008] The Monte Carlo simulation adopted in the present invention is a numerical method based on random sampling and statistical analysis, which is suitable for approximate solutions to high-dimensional and complex problems. Its core steps include defining the model, generating random samples, calculating the output, and statistical analysis. Its advantages are that it is not limited by dimensions and is easy to implement. The use of variance reduction technology and parallel computing can improve its computational efficiency, and verification and correctness checking are also the key to ensuring the reliability of the results. The basic principle of Monte Carlo simulation is to simulate uncertainty by generating a large number of random samples and use statistical results to approximate the problem. The main application scenarios are difficult to obtain analytical solutions, high-dimensional integration, financial derivatives pricing, risk assessment, physical simulation, etc.

[0009] The specific method of the Monte Carlo simulation adopted by the present invention is: First, determine the input variables and their probability distribution (such as uniform distribution, normal distribution, etc.), establish a mathematical relationship between input and output (such as function, differential equation, etc.), use a random number generator (such as pseudo-random number or quasi-random number) to generate samples that conform to the distribution, perform model calculations on each sample, obtain output results (such as option price, integral value, etc.), calculate the output mean, variance, quantile and other statistics, draw a distribution histogram or calculate the confidence interval, and evaluate the reliability of the results.

[0010] The Monte Carlo simulation optimization method adopted in the present invention adopts importance sampling, control variable method, stratified sampling and other methods to reduce variance, and can use multi-core CPU or GPU to accelerate independent samples to achieve parallel computing; the adopted Monte Carlo model transforms deterministic problems into statistical problems through randomness. Its flexibility and adaptability to high-dimensional problems enable it to make fast and accurate predictions for distribution network planning.

[0011] The present invention applies the Monte Carlo model to distribution network planning, mainly based on probability statistics and random simulation methods, to deal with uncertainty problems in planning, and evaluate the risks and performance of different planning schemes through a large number of random sampling. For example, when planning the expansion of the distribution network, the load growth may follow a certain distribution, but the daily output of renewable energy sources such as distributed wind farms and distributed photovoltaic fields is random. Monte Carlo simulation can generate a variety of possible load and renewable energy output scenarios for many years in the future, and then perform flow calculations on each scenario to evaluate whether constraints such as voltage and line capacity are met, and then statistically calculate the reliability and economic indicators of different planning schemes. In addition, it is also necessary to consider the randomness of equipment failures, such as the failure rate of lines or transformers, and use Monte Carlo to simulate random failure events to evaluate the reliability of the system, thereby optimizing equipment configuration or redundant design.

[0012] The present invention performs uncertainty modeling in distribution network planning, mainly using the Monte Carlo method to establish a probability model for the following key uncertainty factors: Load demand: based on historical data or forecasting models (e.g., normal distribution, time series models); Distributed (decentralized) energy: output fluctuations of photovoltaic and wind power (Beta distribution, Weibull distribution); Equipment failure: line and transformer failure rates (Poisson process, exponential distribution); Electricity prices and policies: market electricity price fluctuations, subsidy policy changes (random processes or discrete scenarios).

[0013] The main steps of distribution network planning based on Monte Carlo model include: Step 1: Define planning objectives and constraints: Objective: Minimize investment cost, maximize reliability (e.g., SAIDI), optimize renewable energy penetration, etc. Constraints: Voltage deviation, line capacity, short-circuit current, N-1 criterion, etc.

[0014] Step 2: Generate random scenarios: Generate multiple sets of random parameter combinations (e.g., load curves, wind power output sequences, equipment states) through probability distributions.

[0015] Example: Simulate a 3% ± 1% (normal distribution) annual growth rate in load for the next 10 years, and photovoltaic output following a Beta distribution.

[0016] Step 3: Deterministic planning analysis: For each random scenario, perform the following: Power flow calculation: Verify voltage and power limits; Reliability assessment: Calculate outage frequency and outage time (e.g., Monte Carlo fault simulation); Economic analysis: Calculate investment cost, operation and maintenance cost, and renewable energy benefits.

[0017] Step 4: Statistics and optimization: Indicator statistics: Calculate the mean, variance, and quantiles (e.g., 95% confidence interval) of the objective function (e.g., total cost); Risk assessment: Identify high-risk scenarios (e.g., high load + low renewable energy output leading to overload); Optimization of solutions: Select the most robust planning scheme based on statistical results, or use a stochastic optimization algorithm to obtain the optimal planning scheme (e.g., stochastic programming, chance-constrained programming).

[0018] In embodiments of the present invention, typical application scenarios include: Scenario 1: Distribution network expansion with high proportion of renewable energy, for example: Distributed photovoltaic / distributed wind power output fluctuation leading to voltage limit or reverse power flow; Using the Monte Carlo method, the output curve under different weather conditions can be simulated to evaluate the configuration capacity of energy storage and the expansion demand of lines.

[0019] Scenario 2: Equipment failure and reliability optimization, for example: Line aging leading to increased failure rate, requiring optimization of redundant equipment (e.g., sectionalizing switch, standby transformer); Using the Monte Carlo method, random fault event chains can be simulated to statistically evaluate the power supply reliability (e.g., SAIFI) under different redundant configurations.

[0020] Scenario 3: Electric vehicle charging load planning, for example, the randomness of electric vehicle charging time and power affects the peak load of the distribution network; The Monte Carlo method can be used to generate charging behavior random scenarios, optimize charging pile layout and power grid upgrade scheme.

[0021] Scenario 4: When upgrading the distribution network of a certain city, the Monte Carlo method is used to simulate the load growth and photovoltaic penetration rate in the next 20 years, evaluate the overload probability of different cable selection schemes, and finally select a cost-risk balanced scheme (investment reduced by 15%, overload probability <5%).

[0022] Scenario 5: When optimizing the operation of a microgrid island, simulate wind speed out-of-limit shutdown and distributed photovoltaic output fluctuation, optimize energy storage capacity configuration, and ensure that the key load is continuously powered in 99% of the scenarios.

[0023] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A distribution network planning method based on a Monte Carlo model, characterized by: The planning steps include the following: Step 1: Define the distribution network planning objectives and constraints: Based on the constraints of voltage deviation, line capacity, short-circuit current, and the N-1 criterion, achieve the planning goals of minimizing investment costs, maximizing reliability, and optimizing renewable energy penetration; In distribution network planning, a Monte Carlo model based on probability statistics and stochastic simulation is used to establish a probabilistic model for the following key uncertainties, including: Establish load demand models based on historical data or forecast models; Establish a distributed energy model based on the output fluctuations of photovoltaic and wind power; Establish equipment failure models based on line and transformer failure rates; Establish electricity price and policy models based on market electricity price fluctuations and subsidy policy changes; Step 2: Generate multiple possible future load and renewable energy output scenarios through the Monte Carlo model, and generate multiple sets of random parameter combinations through probability distribution; Step 3: Perform the following processing on each random scene: Power flow calculation: verify whether voltage and power exceed the limit; Reliability assessment: Calculate power outage frequency and duration; Economic analysis: Calculate investment costs, operation and maintenance costs, and renewable energy benefits; Step 4: Calculate and count the mean, variance, and quantile of the objective function, draw a distribution histogram or calculate the confidence interval to identify high-risk scenarios; The planning scheme with the best robustness is selected based on statistical results, or the optimal planning scheme is calculated by combining it with a random optimization algorithm.

2. The distribution network planning method based on the Monte Carlo model according to claim 1, characterized in that: The power flow calculation performed in step 3 specifically includes: performing power flow calculation for each scenario, evaluating whether constraints such as voltage and line capacity are met, and calculating reliability and economic indicators of different planning schemes.

3. The distribution network planning method based on the Monte Carlo model according to claim 1, characterized in that: In step 3, the power outage frequency and power outage time are calculated by using a Monte Carlo model to perform fault simulation.

4. The distribution network planning method based on the Monte Carlo model according to claim 1, characterized in that: The random optimization algorithm used in step 4 is specifically a random programming algorithm or a chance constrained programming algorithm.