Micro-grid power supply reliability evaluation method considering intermittency of renewable energy sources

Abstract

The invention discloses a micro-grid power supply reliability evaluation method considering the intermittency of renewable energy sources. Uncertain data are obtained and modeled as a probability distribution model, and then Latin hypercube sampling and scene reduction are sequentially performed to obtain a key uncertain scene set; modeling the reliability of each component under different output powers according to the relationship between the component failure rate and the output power, and constructing a system reliability model; constructing a reliability risk joint model for the system reliability model by adopting a conditional value-at-risk method; constructing a micro-grid reliability optimization model according to the system reliability model, the reliability risk joint model and the micro-grid operation constraint; and taking the key uncertainty scene set as model input, and solving to obtain an optimized scheduling result and a reliability numerical value corresponding to the optimized scheduling result. According to the method, the accuracy of reliability evaluation is improved, the global optimal solution is ensured, the anti-risk capability of the system is enhanced, and the method is suitable for micro-grid dispatching optimization with high reliability requirements.

Description

Technical Field

[0001] This invention relates to the field of microgrid operation and reliability optimization, and in particular to a method for assessing the reliability of microgrid power supply considering the intermittency of renewable energy. Background Technology

[0002] Microgrids, as an important carrier for achieving efficient and sustainable power distribution, typically have optimization objectives categorized into three types: economic, environmental, and power quality. Among these, power supply reliability is crucial for sensitive load scenarios such as healthcare and education. However, existing technologies often treat reliability indicators as constraints rather than direct optimization targets. Furthermore, traditional reliability indicators (such as Expected Energy Shortage (ENS) and System Average Outage Frequency (SAIFI)) focus on economic losses and are insufficient to dynamically reflect the system's reliability level. Existing technologies suffer from the following shortcomings: 1) The component failure rate is set as a constant, without considering the dynamic impact of output power on the failure rate during actual operation, resulting in inaccurate reliability assessment; 2) Most reliability optimization models are non-convex, which can easily get trapped in local optima and cannot guarantee the optimal global reliability of the system; 3) The intermittency and uncertainty of renewable energy output and load demand can easily trigger worst-case scenario risks, and existing risk assessment methods are difficult to effectively quantify such risks; 4) The balance between reliability and economy is not fully achieved, making it difficult to meet the multi-objective requirements of actual microgrid operation.

[0003] Therefore, there is an urgent need for a microgrid reliability optimization method that can dynamically quantify component reliability, achieve global optimization, resist uncertainty risks, and take into account economic efficiency. Summary of the Invention

[0004] To address the problems in the background art, this invention proposes a microgrid power supply reliability assessment method that takes into account the intermittency of renewable energy.

[0005] The technical solution adopted in this invention is: This invention includes the following steps: S1. Obtain uncertain data, model the uncertain data as a probability distribution model, and then perform Latin hypercube sampling and scene reduction to obtain the key uncertain scene set. S2. Based on the relationship between component failure rate and output power, model the reliability of each component under different output power, and then construct a system reliability model. S3. Use the conditional value at risk method to quantify the risk of the system reliability model and construct a joint reliability risk model; S4. Construct a microgrid reliability optimization model based on the system reliability model, the joint reliability risk model, and the microgrid operation constraints; S5. Using the set of key uncertainty scenarios as input to the microgrid reliability optimization model, solve the microgrid reliability optimization model to obtain the optimized scheduling result and the corresponding reliability value.

[0006] The uncertain data includes uncertainties in photovoltaic irradiance, load demand, and wind speed.

[0007] The scene reduction is specifically handled according to the following formula: in, This represents the value of the uncertain variable obtained from sampling. It represents the inverse function of the distribution function of an uncertain variable. This represents a uniformly divided probability interval. This represents the k-th probability interval. Indicates the number of scenes.

[0008] The relationship between the component failure rate and the output power is specifically set according to the following formula: in, and These represent the minimum and maximum failure rates of the components, respectively. and These represent the minimum and maximum output power of the components, respectively. This is a preset coefficient used to control the overall failure rate level. This is a preset power sensitivity coefficient used to control the failure rate; The system reliability model is specifically set according to the following formula: in, Indicates the overall reliability of the system. This indicates that the system's reliability is assessed in segments. and These represent the failure rates of the upstream power grid and distributed generation, respectively. and These represent the minimum and maximum failure rates of the upstream power grid, respectively. and These represent the minimum and maximum failure rates of the distributed power source, respectively.

[0009] The joint reliability risk model is specifically set according to the following formula: , in, As a preset risk threshold, For loss variables that exceed the risk threshold, For confidence level, For the scene The system reliability objective function value under the following conditions For the scene The probability of.

[0010] Microgrid operation constraints include distributed power generation unit constraints, photovoltaic module output constraints, wind turbine output constraints, and energy storage system constraints.

[0011] The multi-objective function of the microgrid reliability optimization model is specifically set according to the following formula: in, The objective function representing system reliability is... This represents the objective function for the system's operating cost. , and These are the operating costs of distributed power sources, the cost of obtaining electricity from the upstream power grid, and the cost of energy storage losses. For the scene The probability, The confidence level.

[0012] The optimized scheduling results include distributed power generation output plans, energy storage system charging and discharging strategies, upstream grid interaction power plans, and system reliability indicators for each time period.

[0013] The beneficial effects of this invention are: 1) Establish a dynamic correlation model between component failure rate and output power to improve the accuracy of reliability assessment; 2) Transform the non-convex optimization problem into a convex model to ensure that the global optimal solution is obtained; 3) Quantify the worst-case scenario risks brought about by uncertainty to enhance the system's resilience; 4) Achieve multi-objective optimization of reliability and economy, and provide practical microgrid dispatching schemes. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of this invention.

[0017] like Figure 1 As shown, this embodiment includes the following steps: S1. Obtain uncertain data, model the uncertain data as a probability distribution model, and then perform Latin hypercube sampling and scene reduction to obtain the key uncertain scene set. This involves the generation and reduction of uncertainty scenarios. Based on the Latin hypercube sampling (LHS) method, combined with normal and Weibull distributions, a set of uncertainty scenarios for renewable energy output and load is generated, and key scenarios are retained through scenario reduction techniques. S2. Based on the relationship between component failure rate and output power, model the reliability of each component under different output power, and then construct a system reliability model. This involves constructing a power-based failure rate (PFR) model. An exponential correlation model between component failure rate and output power is established to quantify the dynamic impact of output power on component reliability. The nonconvex optimization problem is then linearized. Piecewise linearization is used to handle the nonconvexity of the PFR model, and the product term in the system reliability function is linearized using the McCormick relaxation method, transforming the nonconvex model into a mixed integer linear programming (MILP) model.

[0018] S3. Use the Conditional Value at Risk (CVaR) method to quantify the risk of scenarios where the reliability is below a preset threshold and construct a joint reliability risk model. S4. Construct a microgrid reliability optimization model based on the system reliability model, the joint reliability risk model, and the microgrid operation constraints, and construct a multi-objective function with the optimization objectives of minimizing operating costs and maximizing reliability. This involves modeling the components of a microgrid system, establishing mathematical models for distributed generation (DG), photovoltaic (PV) modules, wind turbines (WT), upstream grid, energy storage systems (ESS), and loads, and clarifying the operational constraints of each component. CVaR-based risk assessment. The Conditional Value at Risk (CVaR) method is introduced to construct a risk assessment model and quantify the impact of worst-case scenarios on system reliability. A multi-objective optimization solution is achieved by combining CVaR-based risk assessment, power-based failure rate (PFR) model, and microgrid system component model. The goal is to maximize reliability and minimize operating costs. e - Combining the constraint method with fuzzy algorithms to solve multi-objective optimization problems and obtain Pareto optimal solutions.

[0019] S5. Using the set of key uncertainty scenarios as input to the microgrid reliability optimization model, solve the microgrid reliability optimization model to obtain the optimized scheduling result and the corresponding reliability value.

[0020] Uncertainty data includes uncertainties in photovoltaic irradiance, load demand, and wind speed. In modeling these uncertainties as probability distributions, photovoltaic irradiance and load demand are modeled as normal probability distributions, while wind speed is modeled as a Weibull probability distribution.

[0021] The method employs Latin hypercube sampling (LHS) to generate an initial scene set based on a normal distribution (describing photovoltaic irradiance and load demand) and a Weibull distribution (describing wind speed), thus solving the problems of low efficiency and insufficient coverage of extreme scenes in traditional Monte Carlo sampling. Scene reduction involves calculating the distance between each scene and the reference scene, setting thresholds to filter key scenes, and retaining high-probability and extreme scenes, thereby reducing computational complexity while ensuring computational accuracy.

[0022] Scene reduction is handled according to the following formula: in, This represents the value of the uncertain variable obtained from sampling. It represents the inverse function of the distribution function of an uncertain variable. This represents a uniformly divided probability interval. This represents the k-th probability interval. Indicates the number of scenes.

[0023] Traditional methods treat component failure rate as a constant. This invention proposes a PFR model to establish an exponential relationship between failure rate and output power. The relationship between component failure rate and output power is specifically set according to the following formula, which can dynamically reflect the impact of output power changes on component reliability, making the evaluation results more realistic: in, and These represent the minimum and maximum failure rates of the components, respectively. and These represent the minimum and maximum output power of the components, respectively. This is a preset coefficient used to control the overall failure rate level. This is a preset power sensitivity coefficient used to control the failure rate; Based on the PFR model, the system reliability function is: Since the non-convex nature of the PFR model makes it difficult to solve, the non-convex optimization problem is linearized. Piecewise linearization is used to handle the non-convex nature of the PFR model, and the McCormick relaxation method is combined to linearize the product term in the system reliability function, thus transforming the non-convex model into a mixed integer linear programming (MILP) model.

[0024] The exponential nonconvexity of the PFR model is transformed into a multi-segment linear function, and linear approximation is achieved by using piecewise slopes and binary selection variables: 1) Power segmentation constraints: in: and These represent the minimum and maximum power of the segments, respectively; This indicates a segmented indicator variable.

[0025] 2) Segmented selection constraints 3) Component output refactoring 4) Failure rate linearization in: Piecewise slope Linearizing the failure rate product term in the system reliability function and relaxing it through upper and lower bound constraints ensures the convexity of the model. The original form of the system reliability function is as follows: Through approximation Transformed into: ,in satisfy: By taking the above two steps, the non-convex model is transformed into a mixed integer linear programming (MILP) model, and the global optimal solution can be obtained through linear solvers such as CLPEX.

[0026] The system reliability model is set according to the following formula: in, Indicates the overall reliability of the system. This indicates that the system's reliability is assessed in segments; and These represent the failure rates of the upstream power grid and distributed power sources, respectively. and These represent the minimum and maximum failure rates of the upstream power grid, respectively. and These represent the minimum and maximum failure rates of the distributed power source, respectively.

[0027] The joint reliability risk model is set up according to the following formula: , in, As a preset risk threshold, For loss variables that exceed the risk threshold, For confidence level, For the scene The objective function value is below. For the scene The probability of.

[0028] Specifically, by introducing the Conditional Value at Risk (CVaR) method, a risk assessment model is constructed to quantify the impact of the worst-case scenario on system reliability; Introducing the Conditional Value at Risk (CVaR) method to quantify the worst-case scenario risk arising from renewable energy and load uncertainty: 1) Defining the risk threshold and loss variables When the scene When the objective function value exceeds the threshold The value is the portion exceeding the limit; otherwise, it is 0. 2) Construct a risk assessment objective function, balancing expected gains and extreme losses, to ensure the microgrid dispatch scheme maintains high reliability even in the worst-case scenario. The final objective function can be expressed as: The constraints are: , In the formula: Risk threshold; For loss variables that exceed the risk threshold; Confidence level; For the scene The objective function value is as follows; For the scene The probability of.

[0029] Microgrid operation constraints include distributed power generation unit constraints, photovoltaic module output constraints, wind turbine output constraints, upstream grid constraints, load constraints, and energy storage system constraints.

[0030] Specifically, the DG unit constraints consider upper and lower limits of output and ramp rate constraints to avoid sudden changes in output affecting system stability. The specific constraints are as follows: In the formula: and These are the upper and lower limits of the DG unit output; For the slope rate constraint 2) The PV module output model is established based on the differences between solar irradiance, ambient temperature, and standard test conditions to calculate the output, reflecting the impact of uncertainties in illumination. The specific constraints of the output model are as follows: In the formula: and PV output and irradiation intensity under standard test conditions; This represents the average solar irradiance. The maximum power temperature coefficient; This indicates the temperature of PV under standard test conditions; Indicates ambient temperature.

[0031] 3) The WT output model divides the output characteristics according to the wind speed range and uses the Weibull distribution to describe the wind speed uncertainty. The specific output model constraints are as follows: In the formula: , , These are the cut-in, rated, and cut-out wind velocities, respectively. Rated power; and This refers to the actual wind speed and actual power output.

[0032] 4) The ESS constraint considers charge / discharge efficiency, energy storage capacity constraints, and charge / discharge mutual exclusion constraints, and implements charge / discharge state control through binary variables: In the formula: Indicates the current capacity of the ESS; and These represent charging power and discharging power, respectively. and These represent charging and discharging indicator variables, respectively. and These represent charging efficiency and discharging efficiency, respectively. and These represent the minimum and maximum states of charge, respectively. Indicates the rated capacity of the ESS; and These represent the maximum charging power and the maximum discharging power, respectively.

[0033] 5) Power balance Finally, a multi-objective optimization solution is performed. The objectives are to maximize reliability and minimize operating cost. e - Combining the constraint method with fuzzy algorithms to solve multi-objective optimization problems and obtain Pareto optimal solutions.

[0034] Economic objectives are transformed into constraints, cost thresholds are set, and reliability objectives are optimized. By adjusting the cost thresholds, a Pareto optimal solution set is generated, and the optimal trade-off point is determined by combining fuzzy algorithms. Operating costs include DG operating costs, upstream grid battery swapping costs, and energy storage loss costs.

[0035] The multi-objective function of the microgrid reliability optimization model is set according to the following formula: Right now in, The objective function representing system reliability is... This represents the objective function for the system's operating cost. , and These are the operating costs of distributed power sources, the cost of obtaining electricity from the upstream power grid, and the cost of energy storage losses.

[0036] The optimized scheduling results include distributed power generation output plans, energy storage system charging and discharging strategies, upstream grid interaction power plans, and system reliability indicators for each time period.

[0037] In summary, this invention achieves a balance between maximizing the annual reliability of microgrids and economic efficiency by constructing a power-based failure rate model, scenario-based stochastic modeling, linearizing non-convex problems, and using conditional value-of-risk (VoV) risk assessment. First, a scenario set containing renewable energy output and load uncertainties is generated based on Latin hypercube sampling. Second, a power-based failure rate model relating component failure rate to output power is established, and the non-convex optimization problem is transformed into a mixed-integer linear programming model through piecewise linearization and McCormick relaxation techniques. Then, the worst-case scenario risk is quantified using the VoV risk assessment method. Finally, through… e - Constraint method for solving multi-objective optimization problems involving reliability and economy.

[0038] The above detailed embodiments illustrate the technical solution and beneficial effects of the present invention. It should be understood that the above description is only the most preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, additions, and equivalent substitutions made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for assessing the reliability of microgrid power supply considering the intermittency of renewable energy sources, characterized in that, The method includes the following steps: S1. Obtain uncertain data, model the uncertain data as a probability distribution model, and then perform Latin hypercube sampling and scene reduction to obtain the key uncertain scene set. S2. Based on the relationship between component failure rate and output power, model the reliability of each component under different output power, and then construct a system reliability model. S3. Use the conditional value at risk method to quantify the risk of the system reliability model and construct a joint reliability risk model; S4. Construct a microgrid reliability optimization model based on the system reliability model, the joint reliability risk model, and the microgrid operation constraints; S5. Using the set of key uncertainty scenarios as input to the microgrid reliability optimization model, solve the microgrid reliability optimization model to obtain the optimized scheduling result and the corresponding reliability value.

2. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 1, characterized in that: The uncertain data includes uncertainties in photovoltaic irradiance, load demand, and wind speed.

3. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 2, characterized in that: The scene reduction is specifically handled according to the following formula: in, This represents the value of the uncertain variable obtained from sampling. It represents the inverse function of the distribution function of an uncertain variable. This represents a uniformly divided probability interval. This represents the k-th probability interval. Indicates the number of scenes.

4. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 1, characterized in that: The relationship between the component failure rate and the output power is specifically set according to the following formula: in, and These represent the minimum and maximum failure rates of the components, respectively. and These represent the minimum and maximum output power of the components, respectively. This is a preset coefficient used to control the overall failure rate level. This is a preset power sensitivity coefficient used to control the failure rate; The system reliability model is specifically set according to the following formula: in, Indicates the overall reliability of the system. This indicates that the system's reliability is assessed in segments. and These represent the failure rates of the upstream power grid and distributed generation, respectively. and These represent the minimum and maximum failure rates of the upstream power grid, respectively. and These represent the minimum and maximum failure rates of the distributed power source, respectively.

5. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 1, characterized in that: The joint reliability risk model is specifically set according to the following formula: ， in, As a preset risk threshold, For loss variables that exceed the risk threshold, For confidence level, For the scene The system reliability objective function value under the following conditions For the scene The probability of.

6. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 1, characterized in that: Microgrid operation constraints include distributed power generation unit constraints, photovoltaic module output constraints, wind turbine output constraints, and energy storage system constraints.

7. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 1, characterized in that: The multi-objective function of the microgrid reliability optimization model is specifically set according to the following formula: in, The objective function representing system reliability is... This represents the objective function for the system's operating cost. , and These are the operating costs of distributed power sources, the cost of obtaining electricity from the upstream power grid, and the cost of energy storage losses. For the scene The probability, The confidence level.

8. The microgrid power supply reliability assessment method considering the intermittency of renewable energy as described in claim 1, characterized in that: The optimized scheduling results include distributed power generation output plans, energy storage system charging and discharging strategies, upstream grid interaction power plans, and system reliability indicators for each time period.

Images