Combined heat and power generation thermal power plant online electricity selling quotation probability curve fitting correction method

Abstract

The invention discloses a cogeneration thermal power plant online electricity selling quotation probability curve fitting correction method, and provides a support reference for an electricity selling enterprise to formulate an electricity selling price strategy for scientifically and reasonably throwing the electricity selling price of the power generation enterprise. According to a probability statistics principle of'bidding Internet surfing and preferential selection 'of provincial power grid dispatching, a bid-winning electricity selling probability curve is obtained through own closest electricity selling historical data and a fitting correction method of a bid-winning electricity price probability curve, and a reference with guiding significance is provided for formulating a weekly pre-electricity selling strategy of a power plant; and the economic value of power selling of the power plant is improved.

Description

Technical Field

[0001] This invention relates to a pricing strategy for electricity sales to cogeneration power plants, and particularly to a method for fitting and correcting the probability curve of electricity sales prices to cogeneration power plants. Background Technology

[0002] Within the State Grid, with the increasing grid connection of new energy sources, the actual output of combined heat and power (CHP) power plants is far lower than their designed output. Currently, CHP power plants sell electricity through competitive bidding in the spot market, resulting in price differences based on time of day and segment. Achieving better economic benefits from electricity sales for CHP power plants has become an urgent goal for these plants.

[0003] The existing pre-sale electricity strategies of various cogeneration power plants are generally conducted through the bidding strategies of cogeneration power spot trading. Currently, the existing bidding strategies for cogeneration power spot trading can be broadly categorized into four types: 1) cost analysis-based methods; 2) market price prediction methods; 3) estimation of bidding behavior by other power generation companies; and 4) game theory-based methods. Among these, the game theory-based method is currently widely used. The game theory-based bidding strategy is designed for the current grid dispatching system's one-time clearing price settlement system for electricity sales. The specific operational steps of the weekly bidding strategy are as follows: each cogeneration power plant submits its on-grid electricity price for each time period of the coming week at the corresponding time point of the week preceding the pre-sale electricity date. After receiving the submitted on-grid pre-sale electricity prices for each time period of the coming week, the provincial grid dispatching center conducts a "bidding" process. The principle of "grid access and selection based on merit" was used to determine the winning cogeneration power plants. The grid-connected electricity prices submitted by each power plant for each power supply period in the coming week were based solely on cost accounting and low-price competition. Because the grid dispatching system did not select the winning bidders based on the lowest price, but rather on a comprehensive consideration of grid power resource allocation and factors such as the power generation capacity, safety, and stability of each power plant, the winning power plants sold electricity at prices far below the grid-side benchmark price given by the provincial dispatching authority. This resulted in the phenomenon of selling electricity at a low price, impacting the economic benefits of the power plants. Therefore, how to scientifically and rationally formulate a power plant's electricity pricing strategy in the week leading up to the pre-sale date to improve its economic efficiency has become a pressing issue for cogeneration power plants in their pre-sale electricity quotations. Summary of the Invention

[0004] This invention provides a method for fitting and correcting the probability curve of electricity sales price for cogeneration thermal power plants, offering a valuable reference for electricity sales companies to formulate scientific and reasonable electricity sales price strategies.

[0005] The present invention solves the above technical problems through the following technical solutions: A method for fitting and correcting the probability curve of electricity sales price at the grid connection of a combined heat and power (CHP) thermal power plant includes the following steps: The first step is to retrieve the benchmark on-grid electricity price Kjz (unit: yuan / megawatt-hour) published by the provincial dispatch authority for the province where the cogeneration thermal power plant is located in the previous year; The second step is to use the time point one week before the pre-sale time of the cogeneration power plant as the base time point, and retrieve all realized one-time clearing prices of electricity sales for the cogeneration power plant in the year prior to the base time point (unit: yuan / megawatt-hour). (If the number of electricity sales days in the year prior to the base time point is d days and the number of electricity sales charges per day is h, then there are a total of d×h one-time clearing prices of electricity sales). The third step is to perform probability statistics on the d×h electricity sales clearing prices obtained in the second step to obtain the distribution curve of the electricity sales clearing price probability Ki, with the horizontal axis being the electricity sales price and the vertical axis being the probability of the electricity sales price occurring. Step 4: Based on the distribution curve of the electricity sales price probability Ki obtained in Step 3, calculate the average value Kμ of all realized one-time clearing price probabilities Ki. The calculation formula is as follows: Kμ=(K1+k2+••••••+Kd×h) / (d×h); Step 5: Calculate the root mean square error σ2 of the electricity sales clearing price probability Ki using the following formula: σ2=∑(Ki-Kμ)2 / (d×h); Step 6: Take the square root of the mean square deviation σ2 from step 5 to obtain the standard deviation σ of the electricity sales clearing price probability Ki. Step 7: Use the standard deviation σ obtained in Step 6 to fit and correct the distribution curve of the power sales clearance price probability Ki obtained in Step 3, so as to obtain the power sales price after one correction for the reference power sales price of cogeneration thermal power plants. Based on the power sales price after one correction probability curve, the cogeneration thermal power plants formulate the power sales price for the next week.

[0006] In the seventh step, the distribution curve of the power sales price probability Ki obtained in the third step is fitted and corrected once using the standard deviation σ. Then, the probability curve after the first correction is fitted and corrected twice using a six sigma table of probability statistics. The cogeneration power plant then formulates the on-grid power sales price for the next week based on the probability curve after the second fitting correction.

[0007] The daily electricity sales billing quantity is 24, 48, or 96 units.

[0008] The proposed on-grid electricity prices for the coming week for cogeneration thermal power plants are all equal to or less than the benchmark on-grid electricity price (Kjz) already published by the provincial dispatch authority for the province where the cogeneration thermal power plant is located in the previous year.

[0009] In the context of today's spot market electricity trading and grid connection, this invention provides a universal reference curve for cogeneration power plants to formulate weekly pre-sale electricity prices. This not only increases the probability of winning the bid but also maximizes the electricity price, enabling the power plant to achieve an electricity price close to the grid connection benchmark price and enhancing the economic value of the power plant's electricity sales. Attached Figure Description

[0010] Figure 1 This is the electricity sales probability fitting correction curve of the present invention; Curve A represents the probability of winning the bid for all realized electricity sales prices in the year prior to the benchmark date for combined heat and power (CHP) thermal power plants; Curve B represents the probability curve after the first adjustment of electricity sales prices; and Curve C represents the probability curve after the second adjustment of electricity sales prices. Detailed Implementation

[0011] The present invention will now be described in detail with reference to the accompanying drawings: Existing weekly pre-sale bids for combined heat and power (CHP) thermal power plants are all based on historical discrete data, and generally, the bids are far lower than the actual grid benchmark price subsequently given by the provincial dispatch center, resulting in a widespread phenomenon of power plants selling electricity at rock-bottom prices. This invention, based on the probabilistic statistical principle of "competitive bidding for grid connection and selection of the best" of the provincial grid dispatch center, uses the plant's closest historical electricity sales data and a fitting and correction method for the probability curve of the winning bid price to derive a probability curve for winning the bid. This provides a guiding reference for the formulation of weekly pre-sale electricity strategies for power plants. The specific steps are as follows: A method for fitting and correcting the probability curve of grid connection electricity sales bids for CHP thermal power plants, including the following steps: The first step is to retrieve the benchmark on-grid electricity price Kjz (unit: yuan / megawatt-hour) published by the provincial dispatch center for the province where the cogeneration power plant is located in the previous year. This benchmark on-grid electricity price Kjz provides the most basic reference price for pre-sale electricity for each cogeneration power plant. The second step is to use the time point one week before the pre-sale date of the cogeneration power plant as the base time point (if the pre-sale date is 0:00 on the first Monday of January, then the base time point is 0:00 on the Monday before the first Monday). Retrieve all the realized one-time clearing prices of electricity sales for the cogeneration power plant in the year prior to the base time point (unit: yuan / megawatt-hour). (If the number of days of electricity sales in the year prior to the base time point is d days and the number of electricity sales charges per day is h, then there are a total of d×h one-time clearing prices for electricity sales). The third step is to perform probability statistics on the d×h electricity clearing prices obtained in the second step to obtain the distribution curve of the probability Ki of the electricity clearing price, with the electricity price on the horizontal axis and the probability of the electricity price occurring on the vertical axis. This statistical work can be performed using existing statistical probability software on a computer to generate the probability curve. Step 4: Based on the distribution curve of the electricity sales price probability Ki obtained in Step 3, calculate the average value Kμ of all realized one-time clearing price probabilities Ki. The calculation formula is as follows: Kμ=(K1+k2+••••••+Kd×h) / (d×h); Step 5: Calculate the root mean square error σ2 of the electricity sales clearing price probability Ki using the following formula: σ2=∑(Ki-Kμ)2 / (d×h); Step 6: Take the square root of the mean square deviation σ2 from step 5 to obtain the standard deviation σ of the electricity sales clearing price probability Ki. This standard deviation σ provides the most basic condition for adjusting historical electricity sales prices from a probabilistic perspective. Step 7: The distribution curve of the power sales price probability Ki obtained in Step 3 is fitted and corrected using the standard deviation σ obtained in Step 6. This involves summing the power sales price probability values ​​obtained in Step 3 with the standard deviation σ, and fitting the resulting set of new power sales price probabilities into a probability curve. This yields the first-corrected probability curve for the power sales price reference of cogeneration power plants. Cogeneration power plants then formulate their on-grid power prices for the following week based on this first-corrected probability curve. Step 7, after first-fitting and correcting the distribution curve of the power sales price probability Ki obtained in Step 3 using the standard deviation σ, further corrects the probability curve using a six-sigma table from probability statistics. Cogeneration power plants then formulate their on-grid power prices for the following week based on the second-corrected probability curve. This second-correction process ensures that the obtained probability curve accurately reflects the actual power sales price from a probabilistic perspective, providing a more scientific reference for the winning bid probability in pre-sale strategies.

[0012] The daily electricity sales quota is 24, 48, or 96 units; corresponding to hourly, 30-minute, and 15-minute electricity sales respectively.

[0013] The proposed on-grid electricity prices for the coming week for cogeneration power plants are all equal to or less than the benchmark on-grid electricity price (Kjz) already published by the provincial dispatch authority for the province where the cogeneration power plant is located in the previous year; removing this portion of the pre-sale electricity price can greatly reduce the risk of not winning the bid.

Claims

1. A method for fitting and correcting the probability curve of electricity sales price at the grid connection of a combined heat and power (CHP) thermal power plant, characterized by the following steps: The first step is to retrieve the benchmark on-grid electricity price Kjz (in yuan / megawatt-hour) published by the provincial dispatch authority for the province where the cogeneration thermal power plant is located in the previous year. The second step is to use the time point one week before the pre-sale time of the cogeneration power plant as the base time point, and retrieve all realized one-time clearing prices of electricity sales for the cogeneration power plant in the year prior to the base time point (unit: yuan / megawatt-hour). (If the number of electricity sales days in the year prior to the base time point is d days and the number of electricity sales charges per day is h, then there are a total of d×h one-time clearing prices of electricity sales). The third step is to perform probability statistics on the d×h electricity sales clearing prices obtained in the second step to obtain the distribution curve of the electricity sales clearing price probability Ki, with the horizontal axis being the electricity sales price and the vertical axis being the probability of the electricity sales price occurring. Step 4: Based on the distribution curve of the electricity sales price probability Ki obtained in Step 3, calculate the average value Kμ of all realized one-time clearing price probabilities Ki. The calculation formula is as follows: Kμ=(K1+k2+••••••+Kd×h) / (d×h); Step 5: Calculate the root mean square error σ2 of the electricity sales clearing price probability Ki using the following formula: σ2 = ∑(Ki - Kμ) 2 / (d×h); Step 6: Take the square root of the mean square deviation σ2 from step 5 to obtain the standard deviation σ of the electricity sales clearing price probability Ki. Step 7: Use the standard deviation σ obtained in Step 6 to fit and correct the distribution curve of the power sales clearance price probability Ki obtained in Step 3, so as to obtain the power sales price after one correction for the reference power sales price of cogeneration thermal power plants. Based on the power sales price after one correction probability curve, the cogeneration thermal power plants formulate the power sales price for the next week.

2. The method for fitting and correcting the probability curve of electricity sales price at a combined heat and power (CHP) thermal power plant according to claim 1, characterized in that, In the seventh step, the distribution curve of the power sales price probability Ki obtained in the third step is fitted and corrected once using the standard deviation σ. Then, the probability curve after the first correction is fitted and corrected twice using a six sigma table of probability statistics. The cogeneration power plant then formulates the on-grid power sales price for the next week based on the probability curve after the second fitting correction.

3. A method for fitting and correcting the probability curve of electricity sales price at a combined heat and power (CHP) thermal power plant according to claim 1 or 2, characterized in that, The daily electricity sales billing quantity is 24, 48, or 96 units.

4. A method for fitting and correcting the probability curve of electricity sales price at a combined heat and power (CHP) thermal power plant according to claim 1 or 2, characterized in that, The proposed on-grid electricity prices for the coming week for cogeneration thermal power plants are all equal to or less than the benchmark on-grid electricity price (Kjz) already published by the provincial dispatch authority for the province where the cogeneration thermal power plant is located in the previous year.

Images