A peak-valley time-of-use electricity price optimization method considering power grid reliability

Abstract

The application discloses a kind of peak valley time-of-use electricity price optimization method considering power grid reliability, the method is aimed at the multi-period electricity price response behavior of user under the environment of electric power market, constructs electricity price elasticity matrix based on the balance relationship of power supply and elastic demand;With the maximization of power grid enterprise income as target, the electricity purchase risk is quantified and is included in objective function, while considering the constraint conditions such as system reliability, line loss and user electricity expense, a peak valley time-of-use electricity price optimization model is established.To improve the calculation efficiency of reliability index in model solving, cubic spline interpolation is introduced to establish the relationship between reliability index and load.The application can significantly improve the power supply reliability of power grid and reduce the electricity purchase risk of power grid enterprise while encouraging users to shave peak and fill valley through price leverage, realizing the win-win of power grid enterprise and user.

Description

Technical Field

[0001] This invention belongs to the intersection of electricity market pricing, demand response, and grid reliability, and specifically relates to a peak-valley time-of-use pricing optimization method that takes grid reliability into account. Background Technology

[0002] With the rapid development of the energy internet and the electricity market, demand-side management (DSM) has become an important means of balancing electricity supply and demand and alleviating peak power supply pressure. Peak-valley time-of-use pricing, as a demand response method based on a price mechanism, guides users to proactively change their electricity consumption behavior by increasing prices during peak hours and decreasing prices during off-peak hours, thereby achieving peak shaving and valley filling.

[0003] However, existing peak-valley time-of-use pricing models have the following significant technical defects and limitations in practical applications:

[0004] 1. Lack of quantitative consideration of the inherent safety (reliability) of the power grid: Traditional models often use "minimizing the mean square error of the load curve" or "minimizing the peak-to-valley difference" as optimization objectives. However, a smooth load curve does not directly equate to improved system reliability. Risk indicators of the power system (such as expected power shortages (EENS)) exhibit a highly nonlinear, exponential relationship with load levels. During periods of extreme peak load, even a small reduction in load can significantly improve reliability; conversely, increasing load during off-peak periods has minimal impact on reliability. Existing models fail to incorporate this nonlinear characteristic into pricing mechanisms.

[0005] 2. Neglecting the risks of purchasing electricity in the spot market: In regions implementing a "single buyer" model or a spot market, power grid companies (electricity sellers) face significant price volatility risks. Spot electricity prices are often positively correlated with system load; that is, the higher the load, the higher the marginal unit cost, and the higher and more volatile the electricity price. Traditional models typically assume fixed electricity purchase costs, ignoring the market characteristic that "peak load = high electricity purchase price + high price volatility risk," which may lead to power grid companies facing huge losses during peak periods due to the formulated electricity pricing scheme.

[0006] 3. Oversimplified User Response Models: Many existing studies only use single-period elasticity matrices (self-elasticity), ignoring the "cross-elasticity" between time periods. In reality, a large portion of the electricity consumption reduced by users during peak hours will be transferred to off-peak and valley periods, rather than disappearing out of thin air. Ignoring cross-elasticity will lead to inaccurate predictions of load transfer amounts, thus affecting the accuracy of power grid dispatching.

[0007] 4. Low computational efficiency of reliability assessment: Traditional reliability assessment (such as Monte Carlo simulation) is extremely time-consuming and difficult to embed directly into the electricity price optimization algorithm loop that requires frequent iterations, resulting in "reliability constraints" often becoming a mere formality or only being able to be verified offline.

[0008] Therefore, there is an urgent need for a peak-valley time-of-use pricing optimization model that can comprehensively consider grid reliability, enterprise electricity purchase risks, and user multi-period response characteristics. Summary of the Invention

[0009] Purpose of the invention: The purpose of this invention is to provide a peak-valley time-of-use pricing optimization method that takes into account grid reliability, enterprise electricity purchase risk and user multi-period response characteristics.

[0010] Technical solution: The present invention provides a peak-valley time-of-use pricing optimization method considering grid reliability, comprising:

[0011] The process involves acquiring the basic load data, retail electricity price parameters, and purchasing electricity market parameters required for electricity price optimization. The basic load data includes typical daily load curve data. The retail electricity price parameters include the current single-rate electricity price, which serves as the initial price for peak, flat, and valley periods. The purchasing electricity market parameters include contract electricity price, contracted electricity volume percentage, and the functional relationship parameters between the average spot electricity price, the standard deviation of the spot electricity price, and the typical daily load level. The functional relationship parameters between the standard deviation of the spot electricity price and the typical daily load level are used to quantify the volatility risk of spot market electricity purchase prices.

[0012] Fuzzy clustering was used to divide the typical daily load curve into three periods: peak, flat, and valley, based on the membership function.

[0013] Based on the linear balance between electricity supply and elastic demand, and using market parameters for peak, flat, and valley periods, the self-elasticity coefficient and cross-elasticity coefficient are derived. The self-elasticity coefficient and cross-elasticity coefficient are used to construct a multi-period response electricity price elasticity matrix. The electricity consumption of users under time-of-use pricing is calculated using the multi-period response electricity price elasticity matrix.

[0014] An analytical functional relationship between typical daily load level and power grid system reliability is established using cubic spline interpolation, resulting in a reliability-load cubic spline function expression. The calculation result of the reliability-load cubic spline function expression is then used as the expected power shortage EENS value under typical load level, thereby constructing a reliability-load cubic spline interpolation model.

[0015] Establish a peak-valley time-of-use electricity pricing optimization model that takes into account grid reliability;

[0016] The constraints of the peak-valley time-of-use pricing optimization model that takes into account grid reliability are established by combining the reliability-load cubic spline interpolation model and the user electricity consumption under the time-of-use pricing. The constraints include system reliability constraints, user electricity fee constraints, price fluctuation constraints, and power balance constraints.

[0017] Using nonlinear programming or intelligent optimization algorithms, under the constraints of the peak-valley time-of-use pricing optimization model that takes into account grid reliability, with the objective of maximizing the expected revenue of grid companies, the optimization model is solved to obtain the optimal peak-hour price, normal-hour price, and valley-hour price. Based on the optimal peak-hour price, normal-hour price, and valley-hour price, the optimization scheme for peak-valley time-of-use pricing that takes into account grid reliability is determined.

[0018] Furthermore, based on the linear balance between electricity supply and elastic demand, and using market parameters for peak, flat, and valley periods to derive self-elasticity coefficients and cross-elasticity coefficients, a multi-period response electricity-price elasticity matrix is ​​constructed using the self-elasticity coefficients and cross-elasticity coefficients. The electricity consumption of users under time-of-use pricing is then calculated using this multi-period response electricity-price elasticity matrix, including:

[0019] Based on economic principles, near the market equilibrium point, during a certain period User electricity consumption With time period electricity price The relationship is approximately linear:

[0020] ;

[0021] In the formula, , , These represent peak hours, normal hours, and valley hours, respectively. For time period The user's electricity consumption; For time period Electricity price; , All are time periods The coefficient of the linear relationship between electricity consumption and electricity price;

[0022] The elements of the elastic matrix are defined as follows:

[0023] ;

[0024] In the formula, Indicates the first The initial electricity price for each time period; Indicates the first The price difference between electricity prices before and after the price adjustment for each time period;

[0025] When r=s, the self-elasticity coefficient is obtained. :

[0026] ;

[0027] When r≠s, the cross elasticity coefficient is obtained. :

[0028] ;

[0029] Electricity-price elasticity matrices for peak, flat, and valley time periods are constructed using self-elasticity coefficients and cross-elasticity coefficients, serving as multi-time period response electricity-price elasticity matrices. :

[0030] ;

[0031] Using the multi-time period response electricity price elasticity matrix Calculate user electricity consumption under time-of-use pricing:

[0032] ;

[0033] in,

[0034] , , ;

[0035] In the formula, This refers to the peak, flat, and valley electricity consumption vectors for users after the implementation of time-of-use pricing. , , These are the optimized electricity consumption during peak, normal, and off-peak hours, respectively. To optimize the user electricity consumption vector for the corresponding time period; , , These measures optimize electricity consumption during peak hours, normal hours, and off-peak hours, respectively. This represents the relative change vector of electricity prices; , , These are the electricity prices for peak hours, normal hours, and off-peak hours, respectively. , , These are the initial electricity prices for the optimized peak, normal, and off-peak periods, respectively.

[0036] Furthermore, the reliability-load cubic spline function expression is as follows:

[0037] Select Load level nodes And the expected power shortage corresponding to each load level node is calculated using a reliability assessment method, denoted as:

[0038] ;

[0039] In the formula, For load level nodes The corresponding EENS value;

[0040] Construct a reliability-load cubic spline function. The general form of the reliability-load cubic spline function is S(L), which is defined over any load subinterval. The reliability-load cubic spline function is expressed as follows:

[0041] ;

[0042] In the formula, For load level Located in the The reliability index value corresponding to each load sub-interval; The load level to be calculated; and For the first The two endpoints of a load sub-interval; , , , For the first The undetermined coefficients of the cubic polynomial within each subinterval.

[0043] Furthermore, the reliability-load cubic spline function needs to satisfy the following interpolation and smoothing conditions:

[0044] Interpolation passed:

[0045] ;

[0046] ;

[0047] First-order continuity:

[0048] ;

[0049] Second-order continuity:

[0050] ;

[0051] Natural boundaries:

[0052] ;

[0053] .

[0054] Furthermore, the objective function of the peak-valley time-of-use pricing optimization model that takes into account grid reliability is as follows:

[0055] ;

[0056] In the formula, This is expressed as the total number of time periods; For the benefit of power grid companies; For the first Time-of-use electricity pricing for different time periods; The first time after the effect of time-of-use pricing Electricity consumption of users in a given time period; For the first The line power loss corresponding to each time period; This represents the percentage of contracted electricity volume. The contracted electricity purchase price; For load level The average spot electricity price under the current conditions; For load level The standard deviation of spot electricity prices below; This is the risk weighting coefficient for electricity purchase.

[0057] Furthermore, the system reliability constraint is expressed as follows:

[0058] ;

[0059] In the formula, For the first Load level for each time period The corresponding EENS value is calculated using the reliability-load cubic spline interpolation model; This is a preset system reliability constraint threshold.

[0060] Furthermore, the user electricity fee constraint is established using the user's electricity consumption under time-of-use pricing, and the expression is as follows:

[0061] ;

[0062] In the formula, For the first Time-of-use electricity pricing for each time period; For the first time after the implementation of time-of-use pricing Electricity consumption of users in a given time period; The original single-rate electricity price; To optimize the first Electricity consumption of users during a specific time period.

[0063] Furthermore, the expression for the price fluctuation constraint is as follows:

[0064] ;

[0065] ;

[0066] And satisfy:

[0067] ;

[0068] In the formula, , , These are the electricity prices for peak hours, normal hours, and off-peak hours, respectively. This is the upper limit of the peak-valley electricity price ratio; , Time periods The lower and upper limits of electricity prices.

[0069] Furthermore, the power balance constraint condition ensures that the total power consumption of users remains unchanged before and after optimization, and its expression is:

[0070] ;

[0071] in, , , These are the optimized electricity consumption during peak, normal, and off-peak hours, respectively. , , The optimizations are for electricity consumption during peak hours, normal hours, and off-peak hours.

[0072] Furthermore, the fuzzy clustering method is used to divide the typical daily load curve into three periods—peak, flat, and trough—based on the membership function, including:

[0073] Suppose a typical day in the middle of the month The load level for each time period is Construct a load sample set The load sample set is clustered using the fuzzy C-means clustering method, with the clustering objective function being... for:

[0074] ;

[0075] in, For the first The load of the time period belongs to the first Membership degree of a class; For fuzzy weighted index, For the first Cluster centers are defined; based on the degree of membership of the three types of cluster centers to the load in each time period, each time period is divided into peak time period, normal time period and valley time period.

[0076] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are as follows:

[0077] (1) By constructing a multi-time period response power price elasticity matrix, this invention incorporates the self-elasticity and cross-elasticity between peak, flat and valley periods of users into the model, thus solving the technical problem in the prior art that only considers single-time period response and is difficult to accurately reflect the load transfer behavior of users, thereby enabling more accurate prediction of load distribution after time-of-use price adjustment.

[0078] (2) This invention establishes a cubic spline interpolation model between load level and reliability index EENS, and introduces the reliability index as an optimization constraint into the peak-valley time-of-use electricity price optimization process. This solves the technical problems in existing methods where load smoothing is not equivalent to reliability improvement and reliability assessment is difficult to embed into the optimization solution, thereby improving the power supply reliability of the system while shaving peaks and filling valleys.

[0079] (3) By incorporating the spot electricity purchase cost and its risk factors into the revenue objective function of the power grid enterprise, this invention solves the technical problem of neglecting the risk of spot electricity purchase price fluctuations during peak hours in the existing time-of-use pricing process, thereby reducing the electricity purchase cost pressure and revenue fluctuation risk of the power grid enterprise during high load periods.

[0080] (4) The present invention uses cubic spline interpolation to replace the repetitive reliability simulation calculation in the optimization iteration process, which solves the technical problem that traditional reliability assessment calculation is inefficient and difficult to use for online or fast optimization solutions, thereby significantly improving the model solution efficiency and engineering application feasibility. Attached Figure Description

[0081] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0082] The technical solution of the present invention will now be described in detail with reference to specific embodiments and accompanying drawings.

[0083] like Figure 1 As shown, the present invention provides a peak-valley time-of-use pricing optimization method that takes into account grid reliability, comprising the following steps:

[0084] S1. Obtain the basic load data, retail electricity price parameters, and purchasing electricity market parameters required for electricity price optimization. The basic load data includes typical daily load curve data; the retail electricity price parameters include the current single-rate electricity price, which serves as the initial electricity price for peak, flat, and valley periods; the purchasing electricity market parameters include the contract electricity price, the contracted electricity volume ratio, and the functional relationship parameters between the average spot electricity price, the standard deviation of the spot electricity price, and the typical daily load level. The functional relationship parameters between the standard deviation of the spot electricity price and the typical daily load level are used to quantify the volatility risk of spot market electricity purchase prices. Specifically:

[0085] Obtain the load for each time period during a typical day This ultimately forms a typical daily load curve, including the system load at different times. The single-rate electricity price includes the initial electricity sales price. Contract electricity price Contracted electricity ratio and average spot electricity price and standard deviation Parameters relating to the load.

[0086] S2. Using fuzzy clustering, the typical daily load curve is divided into three periods: peak, flat, and valley, based on the membership function.

[0087] Multi-time period response electricity price elasticity matrix It is used to accurately calculate the transfer and reduction of electricity caused by changes in electricity prices at different times.

[0088] The specific implementation process of step S2 is as follows:

[0089] S2.1. Using fuzzy clustering, the typical daily load curve is divided into three periods: peak, flat, and valley, based on the membership function. Details are as follows:

[0090] Suppose a typical day in the middle of the month The load level for each time period is Construct a load sample set The load sample set is clustered using the fuzzy C-means clustering method, with the clustering objective function being... for:

[0091] ;

[0092] in, For the first The load of the time period belongs to the first Membership degree of a class; For fuzzy weighted index, For the first Cluster centers are defined; based on the degree of membership of the three types of cluster centers to the load in each time period, each time period is divided into peak time period, normal time period and valley time period.

[0093] S3. Based on the linear balance between electricity supply and elastic demand, and using market parameters for peak, flat, and valley periods, derive the self-elasticity coefficient and cross-elasticity coefficient. Construct a multi-period response electricity price elasticity matrix using the self-elasticity coefficient and cross-elasticity coefficient. Using a multi-time period response electricity price elasticity matrix Calculate the user's electricity consumption under time-of-use pricing.

[0094] The specific implementation process of step S3 is as follows:

[0095] Based on economic principles, near the market equilibrium point, at a certain time... User electricity consumption With time period Electricity price The relationship is approximately linear:

[0096] ;

[0097] In the formula, , , These represent peak hours, normal hours, and valley hours, respectively. For time period The user's electricity consumption; For time period Electricity price; , All are time periods The coefficient of the linear relationship between electricity consumption and electricity price.

[0098] Analyze the user response characteristics to electricity prices and calculate the elements of the elasticity matrix using the following formula:

[0099] The elements of the elastic matrix are defined as follows:

[0100] ;

[0101] In the formula, Indicates the first The initial electricity price for each time period; Indicates the first The price difference between electricity prices before and after the price adjustment for each time period;

[0102] When r=s, the self-elasticity coefficient is obtained. :

[0103] When r=s, the self-elasticity coefficient is obtained. :

[0104] ;

[0105] When r≠s, the cross elasticity coefficient is obtained. :

[0106] ;

[0107] Electricity-price elasticity matrices for peak, flat, and valley time periods are constructed using self-elasticity coefficients and cross-elasticity coefficients, serving as multi-time period response electricity-price elasticity matrices. :

[0108] ;

[0109] Electricity price elasticity matrix using multi-time period response Calculate user electricity consumption under time-of-use pricing:

[0110] ;

[0111] in,

[0112] , , ;

[0113] In the formula, This refers to the peak, flat, and valley electricity consumption vectors for users after the implementation of time-of-use pricing. , , These are the optimized electricity consumption during peak, normal, and off-peak hours, respectively. To optimize the user electricity consumption vector for the corresponding time period; , , These measures optimize electricity consumption during peak hours, normal hours, and off-peak hours, respectively. This represents the relative change vector of electricity prices; , , Electricity prices are categorized into peak, normal, and off-peak hours. , , These are the initial electricity prices for the optimized peak, normal, and off-peak periods, respectively.

[0114] S4. Using cubic spline interpolation, establish the analytical functional relationship between typical daily load level and power grid system reliability, obtain the reliability-load cubic spline function expression, and use the calculation result of the reliability-load cubic spline function expression as the expected power shortage EENS value under typical load level, thereby constructing a reliability-load cubic spline interpolation model.

[0115] Considering the time-consuming nature of directly calculating system reliability indices (such as EENS) using reliability assessment methods like Monte Carlo simulation, this invention pre-calculates the EENS value under typical daily load levels. An analytical functional relationship between typical daily load levels and power grid system reliability is established using cubic spline interpolation, yielding the expression for the reliability-load cubic spline function, as follows:

[0116] Select Load level nodes And the expected power shortage corresponding to each load level node is calculated using a reliability assessment method, denoted as:

[0117] ;

[0118] In the formula, For load level nodes The corresponding EENS value;

[0119] Construct a reliability-load cubic spline function. The general form of the reliability-load cubic spline function is S(L), which is defined over any load subinterval. The reliability-load cubic spline function is expressed as follows:

[0120] ;

[0121] In the formula, For load level Located in the The reliability index value corresponding to each load sub-interval; The load level to be calculated; and For the first The two endpoints of a load sub-interval; , , , For the first The undetermined coefficients of the cubic polynomial within each subinterval.

[0122] The reliability-load cubic spline function needs to satisfy the following interpolation and smoothing conditions:

[0123] (1) Interpolation passed:

[0124] ;

[0125] ;

[0126] (2) First-order continuity:

[0127] ;

[0128] (3) Second-order continuity:

[0129] ;

[0130] (4) Natural boundaries:

[0131] ;

[0132] ;

[0133] Therefore, in the optimization iteration, it is possible to use It performs direct and fast calculations, avoiding Monte Carlo simulations in each iteration.

[0134] This function is used to replace complex reliability simulations in optimization iterations, and returns the system risk value under the current load in milliseconds.

[0135] The reliability-load cubic spline interpolation model is used to quickly obtain the reliability index EENS value in subsequent optimization iterations, which significantly improves the model solution efficiency.

[0136] S5. Establish a peak-valley time-of-use electricity pricing optimization model that takes into account grid reliability; details are as follows:

[0137] The objective is to maximize the revenue of power grid operators. The model considers electricity sales revenue, contract market electricity purchase costs, and spot market electricity purchase costs. In particular, this invention, based on the linear relationship between the mean and variance of spot electricity prices and load, implicitly incorporates control over electricity purchase risk (revenue volatility) in the objective function, i.e., by optimizing the load curve to reduce the expected value and volatility risk of spot electricity purchase costs.

[0138] In this embodiment, the objective function of the peak-valley time-of-use electricity pricing optimization model that takes into account grid reliability (maximizing the expected revenue of the grid company) is as follows:

[0139] ;

[0140] In the formula, This is expressed as the total number of time periods; For the benefit of power grid companies; For the first Time-of-use electricity pricing for different time periods; The first time after the effect of time-of-use pricing Electricity consumption of users in a given time period; For the first The line power loss corresponding to each time period; This represents the percentage of contracted electricity volume. The contracted electricity purchase price; For load level The average spot electricity price under the current conditions; For load level The standard deviation of spot electricity prices below; This is the risk weighting coefficient for electricity purchase.

[0141] S6. Based on the reliability-load cubic spline interpolation model and the user electricity consumption under the time-of-use pricing, establish the constraints of the peak-valley time-of-use pricing optimization model that takes into account grid reliability. The constraints include system reliability constraints, user electricity fee constraints, price fluctuation constraints, and power balance constraints.

[0142] The specific implementation process of step S6 is as follows:

[0143] S6.1 Establish reliability constraints: The optimized annual or periodic EENS value must be less than a set threshold (e.g., a certain percentage of the original EENS).

[0144] The reliability constraints of the system are established using the reliability index EENS value, as shown in the following expression:

[0145] .

[0146] In the formula, For the first Load level for each time period The corresponding EENS value is calculated using the reliability-load cubic spline interpolation model; This is a preset system reliability constraint threshold.

[0147] S6.2 User electricity cost constraints: After the implementation of time-of-use pricing, the total electricity cost for users should not exceed the total electricity cost under the single-rate pricing system, so as to ensure the enthusiasm of users to participate.

[0148] The user electricity fee constraint is established using the user's electricity consumption under time-of-use pricing, and the expression is as follows:

[0149] ;

[0150] In the formula, For the first Time-of-use electricity pricing for each time period; For the first time after the implementation of time-of-use pricing Electricity consumption of users in a given time period; The original single-rate electricity price; To optimize the first Electricity consumption of users during a specific time period.

[0151] S6.3 Price fluctuation constraints: Limit the ratio of peak to off-peak electricity prices (e.g., (Multiple times) is within the set range, and the peak electricity price > flat electricity price > valley electricity price is satisfied. The expression for the price fluctuation constraint is as follows:

[0152] ;

[0153] ;

[0154] And satisfy:

[0155] ;

[0156] In the formula, , , These are the electricity prices for peak hours, normal hours, and off-peak hours, respectively. This is the upper limit of the peak-valley electricity price ratio; , Time periods The lower and upper limits of electricity prices.

[0157] S6.4, Electricity Balance Constraints: Limiting the ratio of peak to off-peak electricity prices (e.g., The power consumption is multiplied by 10 times, ensuring a peak > flat > valley logical relationship. The power balance constraint is that the total power consumption of users remains unchanged before and after optimization, and its expression is:

[0158] ;

[0159] in, , , These are the optimized electricity consumption during peak, normal, and off-peak hours, respectively. , , The optimizations are for electricity consumption during peak hours, normal hours, and off-peak hours.

[0160] S7. Under the constraints of the peak-valley time-of-use pricing optimization model that takes into account grid reliability, with the goal of maximizing the expected revenue of the grid enterprise, solve the peak-valley time-of-use pricing optimization model that takes into account grid reliability to obtain the optimal peak period electricity price, normal period electricity price and valley period electricity price, and determine the peak-valley time-of-use pricing optimization scheme that takes into account grid reliability based on the optimal peak period electricity price, normal period electricity price and valley period electricity price.

[0161] The specific implementation process of step S7 is as follows:

[0162] Based on the solution results, the output includes the optimized electricity pricing scheme, the projected user load curve, the projected reduction in power outage losses (based on EENS improvement), and the revenue increase for the power grid company.

[0163] This invention addresses the multi-period electricity price response behavior of users in a power market environment. It constructs an electricity price elasticity matrix based on the balance between electricity supply and elastic demand. With the goal of maximizing grid company profits, it quantifies electricity purchase risk and incorporates it into the objective function. Simultaneously, it comprehensively considers constraints such as system reliability, line losses, and user electricity expenses to establish a peak-valley time-of-use pricing optimization model. To improve the computational efficiency of reliability indicators in the model solution, cubic spline interpolation is introduced to establish the relationship between reliability indicators and load. This invention can significantly improve grid supply reliability and reduce grid company electricity purchase risk while incentivizing users to smooth peak and valley loads through price leverage, achieving a win-win situation for both grid companies and users.

Claims

1. A method for optimizing peak-valley time-of-use electricity pricing considering grid reliability, characterized in that, include: The process involves acquiring the basic load data, retail electricity price parameters, and purchasing electricity market parameters required for electricity price optimization. The basic load data includes typical daily load curve data. The retail electricity price parameters include the current single-rate electricity price, which serves as the initial price for peak, flat, and valley periods. The purchasing electricity market parameters include contract electricity price, contracted electricity volume percentage, and the functional relationship parameters between the average spot electricity price, the standard deviation of the spot electricity price, and the typical daily load level. The functional relationship parameters between the standard deviation of the spot electricity price and the typical daily load level are used to quantify the volatility risk of spot market electricity purchase prices. Fuzzy clustering was used to divide the typical daily load curve into three periods: peak, flat, and valley, based on the membership function. Based on the linear balance between electricity supply and elastic demand, and using market parameters for peak, flat, and valley periods, the self-elasticity coefficient and cross-elasticity coefficient are derived. The self-elasticity coefficient and cross-elasticity coefficient are used to construct a multi-period response electricity price elasticity matrix. The electricity consumption of users under time-of-use pricing is calculated using the multi-period response electricity price elasticity matrix. An analytical functional relationship between typical daily load level and power grid system reliability is established using cubic spline interpolation, resulting in a reliability-load cubic spline function expression. The calculation result of the reliability-load cubic spline function expression is then used as the expected power shortage EENS value under typical load level, thereby constructing a reliability-load cubic spline interpolation model. Establish a peak-valley time-of-use electricity pricing optimization model that takes into account grid reliability; The constraints of the peak-valley time-of-use pricing optimization model that takes into account grid reliability are established by combining the reliability-load cubic spline interpolation model and the user electricity consumption under the time-of-use pricing. The constraints include system reliability constraints, user electricity fee constraints, price fluctuation constraints, and power balance constraints. Using nonlinear programming or intelligent optimization algorithms, under the constraints of the peak-valley time-of-use pricing optimization model that takes into account grid reliability, with the objective of maximizing the expected revenue of grid companies, the optimization model is solved to obtain the optimal peak-hour price, normal-hour price, and valley-hour price. Based on the optimal peak-hour price, normal-hour price, and valley-hour price, the optimization scheme for peak-valley time-of-use pricing that takes into account grid reliability is determined.

2. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that: The method is based on the linear balance between electricity supply and elastic demand, and derives the self-elasticity coefficient and cross-elasticity coefficient using market parameters for peak, flat, and valley periods. It then constructs a multi-period response electricity-price elasticity matrix using these coefficients, and calculates user electricity consumption under time-of-use pricing using this matrix. This includes: Based on economic principles, near the market equilibrium point, during a certain period User electricity consumption With time period Electricity price The relationship is approximately linear: ； In the formula, , , These represent peak hours, normal hours, and valley hours, respectively. For time period The user's electricity consumption; For time period Electricity price; , All are time periods The coefficient of the linear relationship between electricity consumption and electricity price; The elements of the elastic matrix are defined as follows: ； In the formula, Indicates the first The initial electricity price for each time period; Indicates the first The price difference between electricity prices before and after the price adjustment for each time period; When r=s, the self-elasticity coefficient is obtained. : ； When r≠s, the cross elasticity coefficient is obtained. : ； Electricity-price elasticity matrices for peak, flat, and valley time periods are constructed using self-elasticity coefficients and cross-elasticity coefficients, serving as multi-time period response electricity-price elasticity matrices. : ； Using the multi-time period response electricity price elasticity matrix Calculate user electricity consumption under time-of-use pricing: ； in, ， ， ； In the formula, This refers to the peak, flat, and valley electricity consumption vectors for users after the implementation of time-of-use pricing. , , These are the optimized electricity consumption during peak, normal, and off-peak hours, respectively. To optimize the user electricity consumption vector for the corresponding time period; , , These measures optimize electricity consumption during peak hours, normal hours, and off-peak hours, respectively. This represents the relative change vector of electricity prices; , , These are the electricity prices for peak hours, normal hours, and off-peak hours, respectively. , , These are the initial electricity prices for the optimized peak, normal, and off-peak periods, respectively.

3. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that: The reliability-load cubic spline function expression is as follows: Select Load level nodes And the expected power shortage corresponding to each load level node is calculated using a reliability assessment method, denoted as: ； In the formula, For load level nodes The corresponding EENS value; Construct a reliability-load cubic spline function. The general form of the reliability-load cubic spline function is S(L), which is defined over any load subinterval. The reliability-load cubic spline function is expressed as follows: ； In the formula, For load level L located at the first The reliability index value corresponding to each load sub-interval; The load level to be calculated; and For the first The two endpoints of a load sub-interval; , , , For the first The undetermined coefficients of the cubic polynomial within each subinterval.

4. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 3, characterized in that, The reliability-load cubic spline function needs to satisfy the following interpolation and smoothing conditions: Interpolation passed: ； ； First-order continuity: ； Second-order continuity: ； Natural boundaries: ； 。 5. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that, The objective function of the peak-valley time-of-use electricity pricing optimization model that takes into account grid reliability is as follows: ； In the formula, This is expressed as the total number of time periods; For the benefit of power grid companies; For the first Time-of-use electricity pricing for different time periods; The first time after the effect of time-of-use pricing Electricity consumption of users in a given time period; For the first The line power loss corresponding to each time period; This represents the percentage of contracted electricity volume. The contracted electricity purchase price; For load level The average spot electricity price under the current conditions; For load level The standard deviation of spot electricity prices below; This is the risk weighting coefficient for electricity purchase.

6. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that, The system reliability constraint is expressed as follows: ； In the formula, For the first Load level for each time period The corresponding EENS value is calculated using the reliability-load cubic spline interpolation model; This is a preset system reliability constraint threshold.

7. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that, The user electricity cost constraint is established using the user's electricity consumption under time-of-use pricing, and the expression is as follows: ； In the formula, For the first Time-of-use electricity pricing for each time period; For the first time after the implementation of time-of-use pricing Electricity consumption of users in a given time period; The original single-rate electricity price; To optimize the first Electricity consumption of users during a specific time period.

8. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that, The expression for the price fluctuation constraint is as follows: ； ； And satisfy: ； In the formula, , , These are the electricity prices for peak hours, normal hours, and off-peak hours, respectively. This is the upper limit of the peak-valley electricity price ratio; , Time periods The lower and upper limits of electricity prices.

9. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that, The power balance constraint condition ensures that the total power consumption of users remains unchanged before and after optimization, and its expression is: ； in, , , These are the optimized electricity consumption during peak, normal, and off-peak hours, respectively. , , The optimizations are for electricity consumption during peak hours, normal hours, and off-peak hours.

10. The peak-valley time-of-use pricing optimization method considering grid reliability according to claim 1, characterized in that, The method employs fuzzy clustering to divide a typical daily load curve into three periods—peak, flat, and trough—based on membership functions. Suppose a typical day in the middle of the month The load level for each time period is Construct a load sample set The load sample set is clustered using the fuzzy C-means clustering method, with the clustering objective function being... for: ； in, For the first The load of the time period belongs to the first Membership degree of a class; For fuzzy weighted index, For the first Cluster centers are defined; based on the degree of membership of the three types of cluster centers to the load in each time period, each time period is divided into peak time period, normal time period and valley time period.

Images