Method for evaluating power supply adequacy of electric heating combined system containing new energy and stored energy

Abstract

The invention discloses a power supply adequacy evaluation method for an electric heating combined system containing new energy and energy storage. The method comprises the following steps: acquiring parameters of each thermal power generating unit in the electric heating combined system, historical year load time sequence data, historical year new energy time sequence output data and electric energy storage parameters; obtaining the probability distribution of the net load in each season according to the obtained parameters; obtaining available power generation capacity probability distribution of a thermal power generating unit system according to the parameters of the thermal power generating unit; and carrying out abundance index calculation according to the probability distribution of the net load in each season and the probability distribution of the available power generation capacity of the thermal power generating unit system to obtain an abundance index of the system. According to the system power supply adequacy evaluation method disclosed by the invention, small sample deviation is eliminated, and evaluation robustness is improved; efficient peak load shifting is realized, short visibility of energy storage adjustment is avoided, and a maintenance plan of a unit is reasonably formulated; and finally, the accurate evaluation of the power supply adequacy of the system is realized by quantifying the uncertainty of the source load.

Description

Technical Field

[0001] This invention relates to the field of power system planning and operation reliability assessment technology, and in particular to a method for assessing the power supply adequacy of a combined electric and thermal system containing new energy sources and energy storage. Background Technology

[0002] With the advancement of energy transition, the penetration rate of new energy sources, represented by wind power and photovoltaics, in the power system is continuously increasing. However, the randomness and volatility of new energy output pose significant challenges to the power balance of the power system. Especially in northern heating areas, where combined heat and power (CHP) units are widely used for heating, there is a common "heat-driven power generation" operational constraint, which limits the regulation capacity on the power supply side and further exacerbates the difficulty of the system absorbing new energy sources and ensuring sufficient power supply.

[0003] Power system adequacy assessment is a crucial method for evaluating whether a system has sufficient capacity to meet user demand, considering generator outages and load uncertainties. The mainstream methods for power system adequacy assessment can be broadly categorized into two types: stochastic production simulation and time-series production simulation. Stochastic production simulation, based on probability and statistics theory, does not consider the temporal sequence of load and generator states. Instead, it uses load duration curves and forced outage rates of generators, employing mathematical convolution operations to calculate system reliability indicators and production costs. Time-series production simulation strictly follows a chronological order (typically 8760 hours, or even finer time resolution), simulating the actual operating state of the power system in each time period. It typically combines Monte Carlo simulation or full-time generator combination algorithms to solve for the system's supply and demand balance on a time-by-time basis.

[0004] However, in practical applications, using the above two methods to conduct sufficiency assessments still faces several limitations and challenges, mainly in the following aspects: 1. Time-series production simulations for wind and solar power output often only consider historical data from a single typical year (or a single scenario), making it difficult to comprehensively reflect the interannual fluctuation characteristics of new energy resources over long time scales and the tail risks under extreme climate conditions. Due to the complexity of the climate system, single-year data cannot cover long-term resource change patterns, resulting in assessment results that lack statistical representativeness and are prone to overestimating or underestimating the system's true supply capacity.

[0005] 2. Time-series production simulation faces a computational efficiency bottleneck (curse of dimensionality) with high-dimensional data. When processing 8760 hours of high-resolution data throughout the year, time-series production simulation (especially when combined with sequential Monte Carlo methods) faces enormous computational load and the curse of dimensionality. While hourly simulation across the entire time period can guarantee accuracy, it severely limits solution efficiency. How to achieve efficient probabilistic assessment by reducing the dimensionality of the load and resource seasonal distribution characteristics through reasonable time scales (such as aggregating features from hourly to daily levels) is a pressing technical challenge that needs to be overcome.

[0006] 3. Stochastic production simulations lack refined modeling of the time-series adjustment characteristics of energy storage. These simulations often directly assess the load based on the original load continuity curve, which not only disrupts the temporal continuity of the data but also fails to fully consider the dynamic role of energy storage systems in mitigating new energy fluctuations through peak shaving and valley filling. Because this method ignores the filtering effect of energy storage charging and discharging strategies on extreme load curve fluctuations, it cannot reflect the time-series transfer value of energy storage, ultimately leading to misjudgments of the system's reserve capacity requirements. Summary of the Invention

[0007] This invention addresses the aforementioned problems by proposing a method for evaluating the power supply adequacy of an electrothermal combined system incorporating new energy sources and energy storage. A method for assessing the power supply adequacy of a combined electric and thermal system incorporating new energy sources and energy storage includes the following steps: Step 1: Obtain the parameters of each thermal power unit within the combined electric and thermal power system, and the history of the combined electric and thermal power system. Annual load time series data, historical Annual renewable energy output data and energy storage parameters; Step 2: Based on the parameters of the thermal power unit and the history of the combined electric and thermal power system... Annual load time series data, historical Annual renewable energy output data and energy storage parameters are used to obtain the probability distribution of net load for each season. Step 3: Based on the parameters of the thermal power unit, obtain the probability distribution of the available power generation capacity of the thermal power unit system; Step 4: Calculate the adequacy index based on the probability distribution of net load for each season and the probability distribution of available generating capacity of the thermal power unit system to obtain the system's adequacy index.

[0008] Furthermore, the parameters of the thermal power unit include: installed capacity. Forced shutdown rate and the time required for maintenance ,in , This represents the total installed capacity of thermal power plants in the system. The history Annual load time-series data includes: (The following is a list of data points) Annual electricity load curve , No. Annual heat load curve ,in, The granularity of the time-series data is one hour; The history The annual renewable energy time-series output data is the [number]th The sum of the annual wind power output curve and solar power output curve, expressed as the new energy output curve. It means that, among them, The granularity of the time-series data is one hour; The energy storage parameters include: the maximum energy storage capacity. Maximum charging power Maximum discharge power and the charging / discharging efficiency of electrical energy storage .

[0009] Furthermore, step 2 includes the following steps: Step 21: Based on the history Annual load time series data and the aforementioned history Construction of annual new energy time-series output data The net load curve considering the output of new energy sources is obtained by formula (1): (1) in, The electrical load curve is taken as the first In 2018, the new energy output curve took the first place. The net load curve formed annually after considering the output of new energy sources; Step 22: Establish a variable time-scale rolling filter optimization model, and use the variable time-scale rolling filter optimization model to optimize the filter for different time scales. The net load curve considering renewable energy output, as described in the article, is subjected to variable time-scale rolling filtering to obtain... Net load curve after filtering ; The objective function of the variable time-scale rolling filter optimization model is established as follows: (2) In the formula, To minimize The objective function value of the curve fluctuation; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in Net load value when filtering is performed during the time period; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in The charging power of energy storage during the filtering process; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in The discharge power of stored energy during the filtering process; This represents the total number of time periods in the optimization cycle. Indicates the time period index within the period; Indicates period The average value of the curve over the time period is calculated using the following formula: (3) The constraints of the variable time-scale rolling filter optimization model are established as follows: (4) (5) (6) (7) (8) (9) In the formula: Indicates energy storage The amount of electricity stored during a given period; Indicates energy storage The amount of electricity stored during a given period; Indicates the time granularity, in one hour; Indicates energy storage Discharge power over a given period of time; Indicates energy storage Charging power during a given period; This indicates the initial energy storage capacity for this energy storage cycle; This indicates the amount of energy stored in the last period of the previous optimization cycle; A 0-1 variable, representing electrical energy storage. The power generation status during a given time period is represented by a value of 1, indicating that the energy storage is in a charging state, and a value of 0, indicating that the energy storage is in a discharging state. A 0-1 variable, representing electrical energy storage. The power generation status during a given period is represented by a value of 1, indicating that the energy storage is in a discharging state, and a value of 0, indicating that the energy storage is in a charging state. Net load curve after filtering The calculation formula is as follows: (10); Step 23: Based on the filtered net load curve and heat load curve Calculate the net load after electrothermal decoupling The calculation formula is as follows: (11) This indicates that after considering the output of new energy sources, energy storage filtering, and decoupling of electric and thermal loads, the load that the thermal power units need to bear is obtained. This indicates the power generation loss ratio of thermal power units; Step 24: Establish an optimization model for the maintenance plan of thermal power units based on maximizing and minimizing reserve, and arrange the maintenance of thermal power units using the optimization model to obtain the maintenance periods of the units, based on the net load after electrothermal decoupling. The net load used for statistical distribution is calculated based on the unit's maintenance period. ; The objective function of the thermal power unit maintenance plan optimization model is established as follows: (12) The constraints of the thermal power unit maintenance plan optimization model are established as follows: (13) (14) (15) (16) (17) In the formula: This represents the minimum system reserve capacity during the entire scheduling cycle. For 0-1 state variables, if This indicates thermal power units In the Maintenance must begin immediately, otherwise the value will be 0. This is the set of dates during which maintenance is prohibited, including the peak summer months of July and August, and the peak winter months of January and December; express The maintenance space at any given time is calculated using the following formula: (18) in, For 0-1 state variables, if This indicates thermal power units In the It is always under maintenance; otherwise, the value is 0. Indicates the first The electricity load curve for the year is in the [year] [number]. Load during a specific time period; The net load used for statistical distribution The calculation formula is as follows: (19); Step 25: For the net load used for statistical distribution The probability density distribution characteristics of net load for each season are obtained by performing seasonal and time-period statistical analysis. The specific process is as follows: Using seasons as a dividing line, The net load described in the article for statistical distribution The data is categorized by month, resulting in a net workload dataset containing four independent time sets: spring, summer, autumn, and winter. ; calculate The net load described in the article for statistical distribution global minimum value and global maximum value According to the global minimum value and the global maximum value Set the horizontal axis range for data statistics ; For each seasonal group in Perform the following analysis: The data is statistically analyzed across the horizontal axis. Divide the data evenly into a set number of equal-width intervals and count the results. The net load described in the article for statistical distribution The number of sample points falling into each interval is calculated, and their probability density function estimates are obtained. For the k-th interval, the probability density estimate is... for: (20) in, The total number of samples for that season. The width of each interval, and count is the number of sample points falling within the k-th interval; The kernel density estimation method is used to process the probability density function estimates to obtain the net load probability density function for each season. .

[0010] Furthermore, the probability distribution of available generating capacity of the thermal power unit system in step 3 is calculated as follows: (twenty one) in: Indicates the first The capacity of the Taiwan thermal power unit Indicates the first Forced outage rate of thermal power units in Taiwan Indicates the first The probability of a thermal power unit operating normally. ; For the front The available capacity of the system after combining the two thermal power units is The probability of; for The available capacity of the system after combining the two thermal power units is The probability of; After completing the convolution by traversing all thermal power units using formula (21), the available capacity status table of the thermal power unit system is obtained, which contains a series of discrete capacity values. and their corresponding probability of occurrence For the discrete capacity values and their corresponding probability of occurrence The probability density function of the available generating capacity of the thermal power unit system is obtained by curve fitting. .

[0011] Furthermore, step 4 includes the following steps: Step 41: Based on the probability density function of the available generating capacity of the thermal power unit system, Calculate the cumulative distribution function of the available generating capacity of the thermal power unit system. The calculation formula is as follows: (twenty two) in, For integration variables; Step 42: Based on the cumulative distribution function of the available generating capacity of the thermal power unit system. and the probability density function of net load in each season Calculate the probability of insufficient power The calculation formula is as follows: (twenty three) in, For any given load level; Step 43: Based on the probability density function of the available generating capacity of the thermal power unit system, and the probability density function of net load in each season The calculation process for calculating the expected power shortage is as follows: For a specific load level When the system's power generation capacity At that time, the resulting power deficit is The expected power deficit density at this time for: (twenty four) Combined with the probability density function of net load for this season and the total number of hours in the season Total expected power shortage of thermal power unit system in this season The calculation formula is as follows: (25) (26).

[0012] Furthermore, the optimization cycle in step 22 The calculation formula is as follows: (27) The initial optimization period is 1 day. After the current period's filtering is completed, it is necessary to check whether the energy storage capacity is fully utilized. If the energy storage capacity is not fully utilized, the optimization time span is extended according to formula (27). Until energy storage is fully utilized, among which For a step function, if its independent variable is greater than 0, the function value is 1; otherwise, it is 0. If the first optimization cycle is... If the second round of optimization starts from the hour, then... Start with hourly data and repeat the above steps until you have covered the net load data for 8,760 hours throughout the year.

[0013] Compared with existing technologies, the power supply adequacy assessment method for combined electric and thermal systems incorporating new energy sources and energy storage disclosed in this invention has the following advantages: 1. Comprehensive scenario simulation coverage, overcoming the limitations of single scenarios: This method constructs a panoramic simulation scenario by cross-combining load and renewable energy data from multiple historical years. This not only covers long-term fluctuations in renewable energy output and load, but also makes the evaluation results more reflective of the system's actual operating status under complex and variable conditions.

[0014] 2. The energy storage regulation characterization is refined, solving the quantification challenge under dual constraints: Addressing the dual constraints of energy and power on energy storage systems, this method employs a variable time-scale rolling filter optimization model. This model can finely characterize the time-series regulation characteristics of the energy storage system in mitigating net load fluctuations at different time scales, thereby accurately quantifying its contribution to system adequacy.

[0015] 3. The electrothermal coupling model is accurate and effectively addresses the constraints on unit output: This method, for extraction-condensing cogeneration units, introduces the "power generation loss ratio" index (i.e., the power generation sacrificed for output heat). By converting the heat load into an equivalent electrical load and superimposing it into the base load, it accurately reflects the constraint of heat supply on power generation capacity under electrothermal coupling conditions, thus improving the accuracy of the assessment.

[0016] 4. Scientific and reasonable maintenance considerations improve the overall system reserve level: This method establishes an optimization model for maintenance plans with the goal of maximizing the system's annual minimum reserve capacity. It fully considers the impact of power generation equipment maintenance on system adequacy and effectively avoids the risk of insufficient system adequacy due to improper maintenance arrangements by optimizing maintenance arrangements. Attached Figure Description

[0017] Figure 1 This is a flowchart of the power supply adequacy assessment method for an electric-thermal combined system containing new energy sources and energy storage disclosed in this invention; Figure 2 This is a flowchart of the variable time-scale rolling filtering of the net load curve in this invention; Figure 3 This is the feasible operating domain for the thermal power unit (condensing thermal power unit) in this invention; Figure 4 This is a flowchart of the thermal power unit maintenance plan optimization model in this invention; Figure 5 This is a Gantt chart showing the maintenance schedule for each thermal power unit in this invention; Figure 6 This is a timeline diagram of the annual load and maintenance schedule in this invention; Figure 7 This is a comparison chart of net load distribution in the four seasons in this invention. Detailed Implementation

[0018] like Figure 1 As shown, the method for assessing the power supply adequacy of a combined electric and thermal system incorporating new energy sources and energy storage disclosed in this invention includes the following steps: Step 1: Obtain the parameters of each thermal power unit within the combined electric and thermal power system, and the history of the combined electric and thermal power system. Annual load time series data, historical Annual renewable energy output data and energy storage parameters; Step 2: Based on the parameters of the thermal power unit and the history of the combined electric and thermal power system... Annual load time series data, historical Annual renewable energy output data and energy storage parameters are used to obtain the probability distribution of net load for each season. Step 3: Based on the parameters of the thermal power unit, obtain the probability distribution of the available power generation capacity of the thermal power unit system; Step 4: Calculate the adequacy index based on the probability density distribution characteristics of the net load in each season and the probability distribution of the available generating capacity of the thermal power unit system to obtain the adequacy index of the system.

[0019] The present invention discloses a method for assessing the power supply adequacy of a combined power and heat system incorporating new energy sources and energy storage. This method obtains parameters of each thermal power unit within the combined power and heat system, as well as the historical data of the combined power and heat system. Annual load time series data, historical Annual renewable energy output data and energy storage parameters, based on the parameters of the thermal power unit and the historical data of the combined power and heat system. Annual load time series data, historical By using annual renewable energy output data and energy storage parameters, the probability density distribution characteristics of net load in each season are obtained. Based on the parameters of the thermal power units, the probability distribution of available generating capacity of the thermal power unit system is obtained. Based on the probability distribution of net load in each season and the probability distribution of available generating capacity of the thermal power unit system, adequacy indicators are calculated to obtain the system's adequacy indicators. This invention provides the following beneficial effects for assessing system power supply adequacy: 1. Comprehensive scenario simulation coverage, overcoming the limitations of single scenarios: This method constructs a panoramic simulation scenario by cross-combining load and renewable energy data from multiple historical years. This not only covers long-term renewable energy output fluctuations and load fluctuations but also makes the assessment results more reflective of the system's actual operating status under complex and variable conditions. 2. Refined energy storage regulation characterization, solving the quantification problem under dual constraints: Addressing the dual constraints of energy and power on energy storage systems, this method employs a variable time-scale rolling filter optimization model. This model can finely characterize the time-series regulation characteristics of the energy storage system in smoothing net load fluctuations at different time scales, thereby accurately quantifying its contribution to system adequacy. 3. The electrothermal coupling model is accurate and effectively addresses the constraints on unit output: This method, for extraction-condensing cogeneration units, introduces the "power generation loss ratio" index (i.e., the power generation sacrificed for output heat). By converting the heat load into an equivalent electrical load and adding it to the base load, it accurately reflects the constraint of heating on power generation capacity under electrothermal coupling conditions, improving the accuracy of the assessment. 4. Scientific and reasonable maintenance considerations enhance the overall system reserve level: This method establishes a maintenance plan optimization model aimed at maximizing the system's minimum annual reserve capacity. It fully considers the impact of power generation equipment maintenance on system adequacy, and by optimizing maintenance arrangements, effectively avoids the risk of insufficient short-term system adequacy due to improper maintenance scheduling. Specifically, the detailed steps of the power supply adequacy assessment method for the combined electric and thermal power system containing new energy sources and energy storage disclosed in this application are as follows: Step 1: Obtain the parameters of each thermal power unit within the combined electric and thermal power system, and the history of the combined electric and thermal power system. Annual load time series data, historical Annual renewable energy output data and energy storage parameters; The parameters of the thermal power unit include: installed capacity. Forced shutdown rate and the time required for maintenance ,in , This represents the total installed capacity of thermal power plants in the system. The history Annual load time-series data includes: (The following is a list of data points) Annual electricity load curve , No. Annual heat load curve ,in, The granularity of the time-series data is one hour; The history The annual renewable energy time-series output data is the [number]th The sum of the annual wind power output curve and solar power output curve, expressed as the new energy output curve. It means that, among them, The granularity of the time-series data is one hour; The energy storage parameters include: the maximum energy storage capacity. Maximum charging power Maximum discharge power and the charging / discharging efficiency of electrical energy storage .

[0020] Step 2: Based on the parameters of the thermal power unit and the history of the combined electric and thermal power system... Annual load time series data, historical Annual renewable energy output data and energy storage parameters are used to obtain the probability distribution of net load for each season. Specifically, step 2 includes the following steps: Step 21: Based on the history Annual load time series data and the aforementioned history Construction of annual new energy time-series output data The net load curve considering the output of new energy sources is obtained by formula (1): (1) in, The electrical load curve is taken as the first In 2018, the new energy output curve took the first place. The net load curve formed annually after considering the output of new energy sources; This invention, by considering the cross-combination of multiple historical load curves and renewable energy output curves, can break the specific meteorological-load time-series correlation in a single historical year and construct a system covering different climate years and load growth levels. This method effectively expands the scale of the assessment sample, allowing tail risk events such as "maximum load encountering minimum output" that did not occur in a single data year to be taken into account. This eliminates the assessment bias caused by small sample data and significantly improves the robustness of the adequacy assessment results.

[0021] Step 22: Establish a variable time-scale rolling filter optimization model, and use the variable time-scale rolling filter optimization model to optimize the filter for different time scales. The net load curve considering renewable energy output, as described in the article, is subjected to variable time-scale rolling filtering to obtain... Net load curve after filtering ; Specifically, the overall optimization objective of the variable time-scale rolling filter optimization model is to minimize the fluctuation of the load curve by filtering out larger and smaller values ​​in the load curve through the peak shaving and valley filling effect of energy storage, given the energy storage capacity and the maximum charging and discharging power.

[0022] The objective function of the variable time-scale rolling filter optimization model is established as follows: (2) In the formula, To minimize The objective function value of the curve fluctuation; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in Net load value when filtering is performed during the time period; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in The charging power of energy storage during the filtering process; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in The discharge power of stored energy during the filtering process; This represents the total number of time periods in the optimization cycle. Indicates the time period index within the period; Indicates period The average value of the curve over the time period is calculated using the following formula: (3) Specifically, the objective function of the variable time-scale rolling filter optimization model represented by formula (3) of this invention explicitly incorporates the charging and discharging process of the energy storage system, and automatically achieves peak shaving and valley filling by adjusting the charging and discharging power. The model will prioritize compensation for extreme high loads (peak values) and low loads (valley values), making the utilization of energy storage resources more efficient. Moreover, the objective function adjusts the net load based on the average value of the entire iteration cycle (24 hours, 48 ​​hours or longer), avoiding the model from relying too much on a single adjustment period and being limited to short-term instantaneous demand, thus balancing the charging and discharging operations of energy storage.

[0023] The constraints of the variable time-scale rolling filter optimization model are established as follows: (4) (5) (6) (7) (8) (9) In the formula: Indicates energy storage The amount of electricity stored during a given period; Indicates energy storage The amount of electricity stored during a given period; Indicates the time granularity, in one hour; Indicates energy storage Discharge power over a given period of time; Indicates energy storage Charging power during a given period; This indicates the initial energy storage capacity for this energy storage cycle; This indicates the amount of energy stored in the last period of the previous optimization cycle; A 0-1 variable, representing electrical energy storage. The power generation status during a given time period is represented by a value of 1, indicating that the energy storage is in a charging state, and a value of 0, indicating that the energy storage is in a discharging state. A 0-1 variable, representing electrical energy storage. The power generation status during a given period is represented by a value of 1, indicating that the energy storage is in a discharging state, and a value of 0, indicating that the energy storage is in a charging state. Specifically, formula (4) represents the coupling and transfer relationship of energy storage capacity between adjacent time periods; formula (5) indicates that the energy storage capacity of the first time period of the optimization cycle should be equal to the energy storage capacity of the last time period of the previous optimization cycle, so as to realize the connection and transfer of energy storage capacity; formula (6) represents the energy storage capacity boundary of each time period of energy storage, which is between 0 and the maximum energy storage capacity; formulas (7) and (8) represent the operating boundaries of energy storage during discharge and charging, respectively: during discharge, Take 1, When the value is 0, the energy storage discharge power is between 0 and the maximum discharge power, and the charging power is 0; during charging, Take 1, When the value is 0, the charging power of the energy storage is between 0 and the maximum charging power, and the discharging power is 0. Formula (9) shows that the energy storage cannot be charged and discharged at the same time.

[0024] The energy storage charging and discharging power for each time period is obtained using the variable time-scale rolling filter optimization model established above. and Then, the net load for each period after utilizing energy storage to smooth out fluctuations can be obtained. Net load curve after filtering The calculation formula is as follows: (10); Furthermore, the optimization cycle is performed in this step. The calculation formula is as follows: (27) The specific process is as follows: Figure 2 As shown, the initial optimization period is 24 hours. After the current period's filtering is completed, it is necessary to check whether the energy storage capacity is fully utilized. If the energy storage capacity is not fully utilized, the optimization time span is extended according to formula (27). Until energy storage is fully utilized, among which For a step function, if its independent variable is greater than 0, the function value is 1; otherwise, it is 0. If the first optimization cycle is... If the second round of optimization starts from the hour, then... Start with hourly data and repeat the above steps until you have covered the net load data for 8,760 hours throughout the year.

[0025] Step 23: Based on the filtered net load curve and heat load curve Calculate the net load after electrothermal decoupling The calculation formula is as follows: (11) This indicates that after considering the output of new energy sources, energy storage filtering, and decoupling of electric and thermal loads, the load that the thermal power units need to bear is obtained. This indicates the power generation loss ratio of thermal power units; The conventional unit studied in this invention is an extraction-condensing cogeneration unit, and its feasible operating domain is as follows: Figure 3 As shown. This indicates the power generation loss ratio caused by steam extraction for heating, specifically the amount of power generation that needs to be sacrificed for every unit of heat output by the extraction condensing unit. Its quantitative characteristics are represented by the slope of segment AB in the diagram.

[0026] Step 24: Establish an optimization model for the maintenance plan of thermal power units based on maximizing minimum reserve, and optimize the thermal power units using the optimization model to obtain the minimum net reserve capacity that maximizes the annual capacity. This is based on the net load after electrothermal decoupling. The net load used for statistical distribution is obtained by calculating the minimum net reserve capacity to maximize the whole year. ; To make rational use of the system's maintenance space, a mixed-integer linear programming model was established. The core idea of ​​the model is to maximize the reserve capacity during the system's weakest point of the year (i.e., the day with the lowest reserve capacity) by optimizing the maintenance start time of each unit, while satisfying mandatory constraints such as unit maintenance duration, continuity, and avoiding peak summer / winter demand. The specific algorithm flow is as follows. Figure 4 As shown: The optimization objective of the model is to maximize the system's maintenance margin, i.e., to maximize the minimum net reserve capacity for the entire year. In this invention, the objective function of the thermal power unit maintenance plan optimization model is established as follows: (12) The constraints of the thermal power unit maintenance plan optimization model are established as follows: (13) (14) (15) (16) (17) In the formula: This represents the minimum system reserve capacity during the entire scheduling cycle. For 0-1 state variables, if This indicates thermal power units In the Maintenance must begin immediately, otherwise the value will be 0. This is the set of dates during which maintenance is prohibited, including the peak summer months of July and August, and the peak winter months of January and December; express The maintenance space at any given time is calculated using the following formula: (18) in, For 0-1 state variables, if This indicates thermal power units In the The time is always under maintenance; otherwise, it is 0. Indicates the first The electricity load curve for the year is in the [year] [number]. Load during a specific time period; Specifically, formula (13) indicates that the net reserve capacity of the reserve capacity constraint system each day (total capacity minus load, and then minus the capacity of the units under maintenance that day) must be greater than or equal to the specified value. Formula (14) Maintenance Uniqueness Constraint: Each unit must be scheduled for maintenance only once a year, i.e., the sum of all possible start times is 1. Formula (15) Maintenance Boundary Constraint: Units cannot be scheduled for maintenance at the end of the year or during peak electricity consumption periods when the remaining number of days is insufficient. Maintenance begins at a specific time to ensure it can be completed within a continuous time period. Formula (16) defines the logical constraints between maintenance status and start time, including continuity constraints, if the unit... In the Always under maintenance This indicates that the unit must have been in the past. Maintenance began within a 10-day time window, ensuring continuous and uninterrupted maintenance. Formula (17) prohibits maintenance time constraints, prohibiting the scheduling of unit maintenance during peak grid load months to ensure power supply capacity.

[0027] Given that hourly simulations of 8760 hours per year would lead to a surge in decision variables and trigger the curse of dimensionality, this patent also employs a time-scale dimensionality reduction strategy. This involves extracting daily peak loads as key features, thus consolidating the time granularity from hourly to daily (365 days). This process significantly reduces the computational complexity of the model while preserving key load information.

[0028] The net load used for statistical distribution The calculation formula is as follows: (19); Step 25: For the net load used for statistical distribution The probability density distribution characteristics of seasonal net load are obtained by performing seasonal time-period statistics. The specific process is as follows: To illustrate the seasonal characteristics of the load, this invention selects season as the dividing dimension and extracts... The net load data are categorized by month into four independent time sets: spring (March-May), summer (June-August), autumn (September-November), and winter (December-February), thus forming four net load datasets with distinct seasonal distribution characteristics. In order to conduct probability and statistical analysis by season.

[0029] Using seasons as a dividing line, The net load described in the article for statistical distribution The data is categorized by month, resulting in a net workload dataset containing four independent time sets: spring, summer, autumn, and winter. ; calculate The net load described in the article for statistical distribution global minimum value and global maximum value According to the global minimum value and the global maximum value Set the horizontal axis range for data statistics ; For each seasonal group in Perform the following analysis: The data is statistically analyzed across the horizontal axis. The analysis range is uniformly divided into a predetermined number of equally wide intervals. In this embodiment, the analysis range is unified. Divide into 45 equally wide intervals and perform statistics. The net load described in the article for statistical distribution The number of sample points falling into each interval is calculated, and their probability density function estimates are obtained. For the k-th interval, the probability density estimate is... for: (20) in, The total number of samples for that season. The width of each interval is given, and count is the number of sample points falling within the k-th interval. By counting the number of sample points, a histogram can be formed, which is used to visually display the frequency distribution of the data.

[0030] To obtain smoother and more accurate probability density curves, this invention employs kernel density estimation to process the probability density function estimates and obtain the net load probability density functions for each season. .

[0031] For the merged dataset, the probability density function estimate at any point x is given by the following equation: (28) in: For the kernel function, this method uses the Gaussian kernel function, which has the following form: h represents the bandwidth, and its value affects the smoothness of the density curve. This invention employs an adaptive bandwidth selection strategy, automatically determining the optimal bandwidth based on the standard deviation and sample size of the data. for The i-th sample point in the dataset.

[0032] Step 3: Based on the parameters of the thermal power unit, obtain the probability distribution of the available power generation capacity of the thermal power unit system; Because forced shutdowns of generating units are random, the total available generating capacity of the system is not a fixed value, but a random variable that follows a specific probability distribution. Therefore, this step uses the recursive convolution method to construct a system generating capacity model, aiming to derive the cumulative probability distribution of the system at different capacity levels, laying the foundation for subsequent supply and demand balance analysis.

[0033] For each unit in the system Extract its capacity and forced shutdown rate At this time, the unit The state probability model is expressed as follows: Normal operating state: Capacity is... The probability is Forced shutdown state: capacity is 0, probability is 0. .

[0034] To obtain the probability distribution of the total available capacity of the system This invention starts from the initial state (no unit access) and adds the units one by one to the system model for convolution operation.

[0035] set up For the front The available capacity of the system after combining the units is The probability of adding the first. When the tandem unit is in operation, the probability of the new system state is... Calculated using the following recursive formula: The probability distribution of available generating capacity of the thermal power unit system in step 3 is calculated as follows: (twenty one) in: Indicates thermal power unit Installed capacity Indicates thermal power unit Forced shutdown rate Indicates thermal power unit Probability of normal operating state ; For the front The available capacity of the system after combining the two thermal power units is The cumulative probability; for The available capacity of the system after combining the two thermal power units is The cumulative probability; the first term in formula (21) Representing the The Taiwanese unit experienced a forced shutdown (contributing 0MW), and the total system capacity remained at [value missing]. The situation; the second item Representing the The unit is operating normally (contribution) MW), the total system capacity is from Increase to The situation.

[0036] After completing the convolution by traversing all thermal power units using formula (21), the available capacity status table of the thermal power unit system is obtained, which contains a series of discrete capacity values. and their corresponding probability of occurrence For the discrete capacity values and their corresponding probability of occurrence The probability density function of the available generating capacity of the thermal power unit system is obtained by curve fitting. .

[0037] This yields a complete probability distribution describing the supply-side uncertainties of the system. This distribution quantitatively reflects the system's actual power generation capacity after considering the randomness of unit failures. This result will serve as a core input for subsequent adequacy assessments.

[0038] Step 4: Calculate the adequacy index based on the probability density distribution characteristics of the net load in each season and the probability distribution of the available generating capacity of the thermal power unit system to obtain the adequacy index of the system.

[0039] This step, based on the seasonal net load probability distribution obtained in step 2 and the system available generating capacity probability distribution constructed in step 3, uses the convolution integral method to solve for the system's adequacy index. Specifically, the selected evaluation indicators include the probability of power shortage (LOLP) and the expected power shortage (EENS).

[0040] Specifically, step 4 includes the following steps: No. Each season The net load probability density function is The probability density function of the system's available power generation capacity is: The corresponding cumulative distribution function (CDF) is: .

[0041] Step 41: Based on the probability density function of the available generating capacity of the thermal power unit system, Calculate the cumulative distribution function of the available generating capacity of the thermal power unit system. The calculation formula is as follows: (twenty two) in, For integration variables; Step 42: Based on the cumulative distribution function of the available generating capacity of the thermal power unit system. and the probability density function of net load in each season Calculate the probability of insufficient power The calculation formula is as follows: (twenty three) in, For any given load level; In this embodiment, a numerical integration method (such as the trapezoidal integration method) is used to calculate the above formula. Specifically, the cumulative distribution of power generation capacity is... Interpolation mapping to current seasonal load On the coordinate axis, calculate the system outage risk corresponding to each load point, and sum it with the probability density of load occurrence.

[0042] Step 43: Based on the probability density function of the available generating capacity of the thermal power unit system, and the probability density function of net load in each season The calculation process for calculating the expected power shortage is as follows: The expected power shortage reflects the expected power outage volume due to insufficient power generation capacity within the statistical period.

[0043] For a specific load level When the system's power generation capacity At that time, the resulting power deficit is The expected power deficit density at this time for: (twenty four) Combined with the probability density function of net load for this season and the total number of hours in the season (2190 hours) Total expected power shortage of thermal power unit system during the season The calculation formula is as follows: (25) (26).

[0044] The above calculations enable the quantitative coupling of uncertainties between the source (power generation capacity distribution) and the load (net load distribution), thereby accurately assessing the system power supply reliability level after considering unit maintenance plans and new energy fluctuations.

[0045] This invention discloses a method for assessing the power supply adequacy of a combined electric-thermal system incorporating new energy sources and energy storage. This method constructs a panoramic simulation scenario by cross-combining load and new energy data from multiple historical years to cover long-term fluctuations in both new energy output and load. To address the challenge of quantifying the net load smoothing capability of energy storage systems under dual constraints of energy and power at different time scales, this method employs a variable-time-scale rolling filter optimization model to refine the temporal regulation characteristics of the energy storage system in smoothing net load fluctuations. To consider the output obstruction of thermal power units caused by heating, a power generation loss ratio index is introduced based on the operating characteristics of extraction-condensing thermal power units. This index represents the power generation sacrificed per unit of heat output, and the heat load is converted into an equivalent electrical load and superimposed onto the electrical load. To consider the impact of maintenance plans on system adequacy, this method establishes a maintenance plan optimization model aimed at maximizing the system's minimum annual reserve capacity. Finally, a probability distribution model of power generation capacity and net load is constructed for a specific time period, and the system adequacy index is calculated using the convolution integral method. This accurately assesses the system's power supply reliability level after considering unit maintenance plans and new energy fluctuations.

[0046] A specific embodiment of the present invention is as follows: To verify the effectiveness of the method proposed in this invention, an example of an actual electrothermal combined system operating in a certain region will be used for illustration.

[0047] System Parameters and Data Basis: This system comprises 12 thermal power units (including extraction-condensing cogeneration units), with specific parameters shown in Table 1. The total installed capacity is 3600MW, and the power generation loss ratio due to steam extraction for heating is 0.25. In addition, the system includes 650W of wind power and 200W of photovoltaic power, as well as a supporting energy storage system. The maximum charge / discharge power of the energy storage system is 0.2 times that of the new energy installed capacity, the maximum energy storage capacity is 680MWh, and the charge / discharge efficiency is 98%.

[0048] Table 1 Parameters of Thermal Power Units

[0049] For these 12 thermal power units, maintenance was scheduled according to the season, and the specific results are as follows: Figure 5 and Figure 6 As shown. Following the principle of "avoiding peak hours and utilizing off-peak hours," the maintenance plan aims to balance the system's reserve capacity throughout the year and ensure power supply reliability during critical periods.

[0050] The basic data selected included five years of historical load data and ten years of renewable energy output time-series data for the region as input. The final net load distribution is as follows: Figure 7 As shown. Comparing the four curves, it can be seen that the winter curve exhibits a significant "right skew" and "sharp peak" characteristic, with its probability density peak being the highest (approximately 2.0 × 10⁻⁶). - (³), and concentrated around 3330MW. This indicates that the system operates at high net load for most of the winter, with the minimum net load also maintained above 2400MW, resulting in the greatest pressure to ensure power supply.

[0051] Table 2 Calculation Results

[0052] As shown in Table 2, the system faces the greatest pressure to ensure power supply during winter, with LOLP reaching a high of 0.208 and EENS surging to 25385 MWh, both peak values ​​for the year. The main reason for this phenomenon is the significant thermoelectric coupling effect. In stark contrast, the system operates at its optimal state in summer, with LOLP at only 0.003 and EENS as low as 800 MWh (only about 3.2% of winter's level). This is because the output of new energy sources such as photovoltaic power is typically high in summer, and thermal power units do not need to undertake heating tasks, effectively ensuring system sufficiency. During the transitional seasons, spring benefits from suitable climate and lower load levels, maintaining a low risk level (LOLP at 0.017). However, the risk index rises significantly in autumn (LOLP rises to 0.05), significantly higher than in spring and summer.

[0053] This invention proposes a method for evaluating the power supply adequacy of a combined electric and thermal system incorporating new energy sources and energy storage. Its beneficial effects are mainly reflected in: by constructing... A panoramic simulation scenario effectively captures tail risks such as "extreme load encountering minimum output", eliminating small sample bias and improving assessment robustness; a variable time scale rolling filter model is used to achieve efficient peak shaving and valley filling, avoiding the short-sightedness of energy storage regulation; maintenance plans for the units are reasonably formulated by maximizing minimum reserve through maintenance optimization; and finally, by quantifying source-load uncertainty, an accurate assessment of the system's power supply adequacy is achieved.

[0054] The above description is only a preferred embodiment of the present invention, but 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 scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for assessing the power supply adequacy of a combined electric and thermal system incorporating new energy sources and energy storage, characterized in that, Includes the following steps: Step 1: Obtain the parameters of each thermal power unit within the combined electric and thermal power system, and the history of the combined electric and thermal power system. Annual load time series data, historical Annual renewable energy output data and energy storage parameters; Step 2: Based on the parameters of the thermal power unit and the history of the combined electric and thermal power system... Annual load time series data, historical Annual renewable energy output data and energy storage parameters are used to obtain the probability distribution of net load for each season. Step 3: Based on the parameters of the thermal power unit, obtain the probability distribution of the available power generation capacity of the thermal power unit system; Step 4: Calculate the adequacy index based on the probability distribution of net load for each season and the probability distribution of available generating capacity of the thermal power unit system to obtain the system's adequacy index.

2. The method for assessing the power supply adequacy of a combined electric and thermal system containing new energy sources and energy storage as described in claim 1, characterized in that: The parameters of the thermal power unit include: installed capacity. Forced shutdown rate and the time required for maintenance ,in , This represents the total installed capacity of thermal power plants in the system. The history Annual load time-series data includes: (The following is a list of data points) Annual electricity load curve , No. Annual heat load curve ,in, The granularity of the time-series data is one hour; The history The annual renewable energy time-series output data is the [number]th The sum of the annual wind power output curve and solar power output curve, expressed as the new energy output curve. It means that, among them, The granularity of the time-series data is one hour; The energy storage parameters include: the maximum energy storage capacity. Maximum charging power Maximum discharge power and the charging / discharging efficiency of electrical energy storage .

3. The method for assessing the power supply adequacy of an electrothermal combined system containing new energy sources and energy storage according to claim 2, characterized in that: Step 2 includes the following steps: Step 21: Based on the history Annual load time series data and the aforementioned history Construction of annual new energy time-series output data The net load curve considering the output of new energy sources is obtained by formula (1): （1） in, The electrical load curve is taken as the first In 2018, the new energy output curve took the first place. The net load curve formed annually after considering the output of new energy sources; Step 22: Establish a variable time-scale rolling filter optimization model, and use the variable time-scale rolling filter optimization model to optimize the filter for different time scales. The net load curve considering renewable energy output, as described in the article, is subjected to variable time-scale rolling filtering to obtain... Net load curve after filtering ; The objective function of the variable time-scale rolling filter optimization model is established as follows: （2） In the formula, To minimize The objective function value of the curve fluctuation; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in Net load value when filtering is performed during the time period; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in The charging power of energy storage during the filtering process; Indicates the selection of the electrical load curve. Article 1, Selection of the new energy output curve The load curve formed by the bars, in The discharge power of stored energy during the filtering process; This represents the total number of time periods in the optimization cycle. Indicates the time period index within the period; Indicates period The average value of the curve over the time period is calculated using the following formula: （3） The constraints of the variable time-scale rolling filter optimization model are established as follows: （4） （5） （6） （7） （8） （9） In the formula: Indicates energy storage The amount of electricity stored during a given period; Indicates energy storage The amount of electricity stored during a given period; Indicates the time granularity, in one hour; Indicates energy storage Discharge power over a given period of time; Indicates energy storage Charging power during a given period; This indicates the initial energy storage capacity for this cycle. This indicates the amount of energy stored in the last period of the previous optimization cycle; A 0-1 variable, representing electrical energy storage. The power generation status during a given time period is represented by a value of 1, indicating that the energy storage is in a charging state, and a value of 0, indicating that the energy storage is in a discharging state. A 0-1 variable, representing electrical energy storage. The power generation status during a given period is represented by a value of 1, indicating that the energy storage is in a discharging state, and a value of 0, indicating that the energy storage is in a charging state. Net load curve after filtering The calculation formula is as follows: （10）； Step 23: Based on the filtered net load curve and heat load curve Calculate the net load after electrothermal decoupling The calculation formula is as follows: （11） This indicates that after considering the output of new energy sources, energy storage filtering, and decoupling of electric and thermal loads, the load that the thermal power units need to bear is obtained. This indicates the power generation loss ratio of a thermal power unit; Step 24: Establish an optimization model for the maintenance plan of thermal power units based on maximizing and minimizing reserve, and arrange the maintenance of thermal power units using the optimization model to obtain the maintenance periods of the units, based on the net load after electrothermal decoupling. The net load used for statistical distribution is calculated based on the unit's maintenance period. ; The objective function of the thermal power unit maintenance plan optimization model is established as follows: （12） The constraints of the thermal power unit maintenance plan optimization model are established as follows: （13） （14） （15） （16） （17） In the formula: This represents the minimum system reserve capacity during the entire scheduling cycle. For 0-1 state variables, if This indicates thermal power units In the Maintenance must begin immediately; otherwise, the value is 0. The set of dates during which maintenance is prohibited includes the peak summer months of July and August, and the peak winter months of January and December; express The maintenance space at any given time is calculated using the following formula: （18） in, For 0-1 state variables, if This indicates thermal power units In the It is always under maintenance; otherwise, the value is 0. Indicates the first The electricity load curve for the year is in the [year] [number]. Load during a specific time period; The net load used for statistical distribution The calculation formula is as follows: （19）； Step 25: For the net load used for statistical distribution The probability distribution of net load for each season is obtained by performing seasonal and time-period statistics. The specific process is as follows: Using seasons as a dividing line, The net load described in the article for statistical distribution The data is categorized by month, resulting in a net workload dataset containing four independent time sets: spring, summer, autumn, and winter. ; calculate The net load described in the article for statistical distribution global minimum value and global maximum value According to the global minimum value and the global maximum value Set the horizontal axis range for data statistics ; For each seasonal group in Perform the following analysis: The data is statistically analyzed across the horizontal axis. Divide the data evenly into a set number of equal-width intervals and count the results. The net load described in the article for statistical distribution The number of sample points falling into each interval is calculated, and their probability density function estimates are obtained. For the k-th interval, the probability density estimate is... for: （20） in, The total number of samples for that season. The width of each interval, and count is the number of sample points falling within the k-th interval; The kernel density estimation method is used to process the probability density function estimates to obtain the net load probability density function for each season. .

4. The method for assessing the power supply adequacy of an electrothermal combined system containing new energy sources and energy storage according to claim 3, characterized in that: The probability distribution of available generating capacity of the thermal power unit system in step 3 is calculated as follows: （21） in: Indicates the first The installed capacity of Taiwan's thermal power units Indicates the first Forced outage rate of thermal power units in Taiwan Indicates the first The probability of normal operation of a thermal power unit. ; For the front The available capacity of the system after combining the two thermal power units is The probability of; for The available capacity of the system after combining the two thermal power units is The probability of; After completing the convolution by traversing all thermal power units using formula (21), the available capacity status table of the thermal power unit system is obtained, which contains a series of discrete capacity values. and their corresponding probability of occurrence For the discrete capacity values and their corresponding probability of occurrence The probability density function of the available generating capacity of the thermal power unit system is obtained by curve fitting. .

5. The method for assessing the power supply adequacy of an electrothermal combined system containing new energy sources and energy storage according to claim 4, characterized in that: Step 4 includes the following steps: Step 41: Based on the probability density function of the available generating capacity of the thermal power unit system, Calculate the cumulative distribution function of the available generating capacity of the thermal power unit system. The calculation formula is as follows: （22） in, For integration variables; Step 42: Based on the cumulative distribution function of the available generating capacity of the thermal power unit system. and the probability density function of net load in each season Calculate the probability of insufficient power The calculation formula is as follows: （23） in, For any given load level; Step 43: Based on the probability density function of the available generating capacity of the thermal power unit system, and the probability density function of net load in each season The calculation process for calculating the expected power shortage is as follows: For a specific load level When the system's power generation capacity At that time, the resulting power deficit is The expected power deficit density at this time for: （24） Combined with the probability density function of net load for this season and the total number of hours in the season Total expected power shortage of thermal power unit system in this season The calculation formula is as follows: （25） （26）。 6. The method for assessing the power supply adequacy of an electrothermal combined system containing new energy sources and energy storage as described in claim 5, characterized in that: Optimization cycle in step 22 The calculation formula is as follows: （27） The initial optimization period is 24 hours. After the current period's filtering is completed, it is necessary to check whether the energy storage capacity is fully utilized. If the energy storage capacity is not fully utilized, the optimization time span is extended according to formula (27). Until energy storage is fully utilized, among which For a step function, if its independent variable is greater than 0, the function value is 1; otherwise, it is 0. If the first optimization cycle is... If the second round of optimization starts from the hour, then... Start with hourly data and repeat the above steps until you have covered the net load data for 8,760 hours throughout the year.

Images