Urban power demand prediction method based on Logistic model

Abstract

The invention relates to the technical field of power prediction, in particular to an urban power demand prediction method based on a Logistic model. According to the method, the daily power consumption and the power load change trend are analyzed, the overall expansion continuous expression degree, the continuous high load expression degree and the load decline expression degree of each power supply area are obtained, and a traditional industrial area and a novel load area are screened out; obtaining support times and supported times of different power supply areas, and obtaining a comprehensive expansion dynamic influence degree of each novel load area in combination with the overall expansion continuous expression degree of each novel load area; and for the latest historical year, obtaining an electric power saturation capacity influence coefficient, obtaining an optimized saturation capacity in combination with distribution of electrical loads in different historical years, and predicting the urban electric power demand. According to the method, the accurate saturation capacity during Logistic model prediction is obtained, so that the accuracy of power demand prediction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power forecasting, and in particular to a method for forecasting urban power demand based on a Logistic model. Background Art

[0002] Urban electricity demand forecasting is crucial for ensuring stable power supply, optimizing energy resource allocation, reducing system operating costs, improving environmental sustainability, addressing climate change, supporting the development of smart grids, and promoting long-term urban development. Power demand forecasting not only helps improve power system efficiency but also provides a crucial basis for environmental protection and urban infrastructure development, ensuring that grid expansion keeps pace with urban expansion.

[0003] In existing technologies, electricity demand is predicted using a logistic model based on a fixed saturated capacity value. However, the continuous expansion and upgrading of cities has caused the growth of electricity demand to be highly nonlinear and time-varying. The fixed saturated capacity value fails to reflect dynamic changes, resulting in a deviation between the prediction results and actual demand, and the electricity demand prediction effect is poor. Summary of the Invention

[0004] In order to solve the technical problem that fixed saturation capacity values ​​fail to reflect dynamic changes and the power demand forecasting effect is poor, the purpose of the present invention is to provide a method for predicting urban power demand based on the Logistic model. The technical solution adopted is as follows: The present invention proposes a method for predicting urban power demand based on a logistic model, the method comprising: Obtain the daily power consumption of different power supply areas in the city, as well as the power load at each time of the day; Based on the distribution of daily power consumption in the power supply area on all days of the year and the power load change trend in the same period of different days, the overall expansion sustainability performance, sustained high load performance, and load decline performance of each power supply area are obtained; Based on the changes in daily power consumption across all days of the year in the power supply area, traditional industrial areas and new load areas are screened out. The number of times each power supply area supports and is supported is obtained, and combined with the overall expansion sustainability performance of each new load area, the comprehensive expansion dynamic impact of each new load area is obtained. For the latest historical year, the power saturation capacity impact coefficient is obtained based on the sustained high load performance of different power supply areas, the load decline performance of all traditional industrial areas, and the comprehensive expansion dynamic impact of all new load areas. The optimized saturation capacity is obtained based on the power saturation capacity impact coefficient and the distribution of power load in different years. The urban electricity demand is predicted based on the Logistic model after optimizing the saturated capacity.

[0005] Furthermore, the method for obtaining the overall expansion continuous performance includes: In the neighborhood of the latest historical year, the accumulated daily power consumption of each power supply area on all days of each month is obtained and normalized to serve as the power consumption weight of each power supply area in each month. Obtain the difference in power consumption weights between different months and the previous month as the weight difference; consider all months with consecutive positive weight differences to constitute an expansion development cycle; calculate the mean of the power consumption weights of all months in each expansion development cycle as the expansion power consumption weight degree; calculate the mean of the ratio of the cumulative daily power consumption between different months and the previous month in each expansion development cycle as the expansion coefficient; The expansion power consumption weight is gain-adjusted according to the expansion coefficient of each expansion development cycle as the local expansion sustainability performance of each expansion development cycle; the local expansion sustainability performance with the largest value in all expansion development cycles is selected as the overall expansion sustainability performance of each power supply area.

[0006] Furthermore, the method for obtaining the sustained high load performance includes: Based on the daily power consumption distribution of each power supply area on all days of the year, the annual power consumption continuity of each power supply area is obtained; Obtain the power load curves fitted by the power load of each power supply area at the same time period on different days. Select the two power load curves with the highest and lowest positions in all time periods. Calculate the area difference between the two curves through definite integral and perform negative correlation normalization to obtain the multi-period continuous load degree of each power supply area. The annual continuity of power consumption is gain-adjusted according to the multi-period continuous load of each power supply area to obtain the continuous high load performance of each power supply area.

[0007] Furthermore, the method for obtaining the annual continuity of power consumption includes: Normalize the daily power consumption of each power supply area to serve as the daily power consumption weight of each power supply area; Obtain a fitting curve for the power consumption weights of each power supply area on all days, obtain the maximum point of the fitting curve, and obtain the cumulative value of the difference between the power consumption weights of different maximum points and the previous maximum point as the annual power consumption continuity of each power supply area.

[0008] Furthermore, the method for obtaining the load decay performance includes: Obtain the average power load value of each power supply area in all time periods of each day as the average power load level; obtain the minimum point in the curve composed of the average power load levels of all days of the year as the load decay point; According to the average power load level distribution of each load decay point on the left and right sides, the decay node performance of each load decay point is obtained; The attenuation node with the largest performance degree among all load attenuation points is selected, and the attenuation node performance degree of the corresponding load attenuation point is used as the load attenuation performance degree of each power supply area.

[0009] Furthermore, the method for obtaining the attenuated node performance includes: The average power load level corresponding to the load decay point is taken as the target load. The first decay degree is obtained according to the first difference between the maximum value of the average power load level in all days and the target load, and the second difference between the average values ​​of all the average power load levels on the left and right sides of each load decay point. Both the first difference and the second difference are positively correlated with the first decay degree. Obtain the range of consecutive days on the left side during which the average power load level is greater than or equal to the target load as the first time interval; obtain the range of consecutive days on the right side during which the average power load level is less than or equal to the target load as the second time interval; obtain the second decay degree based on the first difference and the interval difference between the second time interval and the first time interval, where the first difference is negatively correlated with the second decay degree, and the interval difference is positively correlated with the second decay degree; The sum of the first attenuation degree and the second attenuation degree of each load reduction point is obtained as the attenuation node representation degree of each load reduction point.

[0010] Furthermore, the method for obtaining the traditional industrial area and the new load area includes: Obtain the sum of the daily power consumption of each power supply area on all days of each year as the overall power consumption of each power supply area in each year; Calculate the difference in overall power consumption between the latest historical year and the previous year for each power supply area as the power consumption decline change degree of each power supply area; If the power consumption decay variation degree of the power supply area is less than or equal to the preset decay threshold, the corresponding power supply area is regarded as a traditional industrial area; otherwise, the corresponding power supply area is regarded as a new load area.

[0011] Furthermore, the method for obtaining the comprehensive expansion dynamic influence includes: For each power supply area, the ratio between the number of support times and the number of supported times is obtained as the power contribution; According to the number of times the adjacent areas of each new load area support the new load area, the number of times the new load area is supported, the power contribution and the overall expansion sustainability performance, the comprehensive expansion dynamic influence of each new load area is obtained. The number of times the adjacent areas support the new load area and the power contribution are negatively correlated with the comprehensive expansion dynamic influence, while the number of times the new load area is supported and the overall expansion sustainability performance are positively correlated with the comprehensive expansion dynamic influence.

[0012] Furthermore, the method for obtaining the power saturation capacity influence coefficient includes: Obtaining the product of the sustained high load performance and the load decay performance of all traditional industrial areas as the first performance value; Obtaining a product cumulative value of the sustained high load performance degree and the comprehensive expansion dynamic impact degree of all new load areas as a second performance value; A difference between the first performance value and the second performance value is obtained as a power saturation capacity influence coefficient.

[0013] Furthermore, the method for obtaining the optimized saturation capacity includes: Obtain the maximum value of the electricity load in a preset number of historical years, use the nonlinear least squares method to fit the logistic curve, and obtain the initial saturation capacity of the latest historical year; The initial saturation capacity is gain-adjusted according to the power saturation capacity influence coefficient to obtain the optimized saturation capacity.

[0014] The present invention has the following beneficial effects: The present invention analyzes the trend of daily power consumption and power load changes to obtain the overall expansion sustainability performance, sustained high load performance, and load decline performance of each power supply area, reflecting the trend change characteristics of regional power consumption. Based on the daily power consumption changes of the power supply area on all days of different years, traditional industrial areas and new load areas are screened out, which is conducive to dividing regional types for targeted analysis. The number of support and supported times of each power supply area is obtained. The number of support and supported times of different power supply areas is obtained, and combined with the overall expansion sustainability performance of each new load area, the comprehensive expansion dynamic impact of each new load area is obtained to evaluate the dynamic impact of new loads on load expansion. For the latest historical year, based on the sustained high load performance of different power supply areas, the load decline performance of all traditional industrial areas, and the comprehensive expansion dynamic impact of all new load areas, the power saturation capacity influence coefficient is obtained to determine the dynamic impact coefficient of the power saturation capacity change of the entire city. Based on the power saturation capacity influence coefficient and the distribution of power load in different years, the optimized saturation capacity is obtained to more accurately predict the power demand inflection point, and the city's power demand is predicted. The present invention improves the accuracy of power demand prediction by obtaining accurate saturation capacity when predicting using a logistic model. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flow chart of a method for predicting urban power demand based on a Logistic model provided by one embodiment of the present invention; Figure 2 A flow chart of a method for obtaining overall expansion persistence performance provided by one embodiment of the present invention; Figure 3 A flowchart of a method for obtaining attenuated node expressiveness provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a method for predicting urban power demand based on a logistic model proposed by the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following describes in detail a specific solution of a method for predicting urban power demand based on a Logistic model provided by the present invention with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of a method for predicting urban power demand based on a logistic model provided by an embodiment of the present invention. The specific method includes: Step S1: Obtain the daily power consumption of different power supply areas in the city and the power load in each time period of each day.

[0021] In an embodiment of the present invention, accurate power demand forecasting requires analysis of power consumption and load. First, a city's power grid GIS platform is used to obtain information about different power supply areas within the city. The power consumption and load data for each area are then collected using a power information collection system and a SCADA system. This information is used to determine the daily power consumption and load for each power supply area within the city, as well as the power load for each time period.

[0022] It should be noted that, in the embodiment of the present invention, each hour is regarded as a time period.

[0023] Step S2: Based on the daily power consumption distribution of the power supply area on all days of the year and the power load change trend in the same period of different days, the overall expansion sustainability performance, sustained high load performance and load decline performance of each power supply area are obtained.

[0024] Different power supply areas in a city have their own power load impact performance. The more continuously the daily power consumption increases, the greater the expansion sustainability performance. The power load change trend in the same period of time on different days reflects the convergence of the curves in multiple periods. The closer the power load performance in multiple periods is to each other, the more obvious the multi-period sustainability performance is. If there are major changes in the power supply area, different power load trends will appear before and after a certain time node, reflecting the degree of load decline performance. Therefore, based on the daily power consumption distribution of the power supply area on all days of the year, and the power load change trend in the same period of time on different days, the overall expansion sustainability performance, sustained high load performance, and load decline performance of each power supply area are obtained.

[0025] Preferably, in one embodiment of the present invention, the method for obtaining the overall expansion continuous performance degree can be found in Figure 2 , which shows a flow chart of a method for obtaining overall expansion continuous performance, including: Step S201: within the neighborhood range of the latest historical year, obtain the accumulated daily power consumption values ​​of each power supply area on all days of each month, and normalize them to serve as the power consumption weight of each power supply area in each month.

[0026] It should be noted that, in one embodiment of the present invention, the neighborhood range of the latest historical year is based on the latest historical year and is composed of the range of the previous year. In other embodiments of the present invention, the size of the neighborhood range can be set according to specific circumstances, and is not limited or elaborated here.

[0027] It should be noted that, in the embodiment of the present invention, linear normalization or a normalization function is used for normalization, and the specific means are technical means well known to those skilled in the art and will not be described in detail here.

[0028] Step S202: Obtain the difference in power consumption weights between different months and the previous month as the weight difference; constitute an expansion development cycle with all months corresponding to continuous positive weight differences; calculate the average of the power consumption weights of all months in each expansion development cycle as the expansion power consumption weight degree; calculate the average of the ratios of the cumulative daily power consumption values ​​between different months and the previous month in each expansion development cycle as the expansion coefficient.

[0029] The greater the increase in electricity load, the more expansionary it is. Therefore, all months with positive continuous weight differences constitute an expansion development cycle for analysis; the expansion power consumption weight within the expansion development cycle is quantified by taking the average value to reflect the overall level of power consumption weight within the expansion development cycle; the greater the cumulative daily power consumption value of the latter month is than that of the previous month, the greater the growth in electricity load and the greater the expansion coefficient.

[0030] Step S203: Gain-adjust the expansion power consumption weight according to the expansion coefficient of each expansion development cycle to serve as the local expansion sustainability performance of each expansion development cycle; select the local expansion sustainability performance with the largest value in all expansion development cycles as the overall expansion sustainability performance of each power supply area.

[0031] It should be noted that the larger the expansion coefficient, the greater the expansion power consumption weight, the greater the growth of electricity load, and the greater the local expansion sustainability performance; the gain adjustment method is: obtain the sum of the positive integer 1 and the expansion coefficient as the expansion adjustment coefficient; calculate the product of the expansion adjustment coefficient and the expansion power consumption weight as the local expansion sustainability performance.

[0032] Preferably, in one embodiment of the present invention, the method for obtaining the sustained high load performance includes: Based on the daily power consumption distribution of each power supply area on all days of the year, the annual power consumption continuity of each power supply area is obtained; Preferably, in one embodiment of the present invention, the method for obtaining the annual continuity of power consumption includes: Normalize the daily power consumption of each power supply area to serve as the daily power consumption weight of each power supply area; Obtain a fitting curve for the power consumption weights of each power supply area on all days, obtain the maximum point of the fitting curve, and obtain the cumulative value of the difference between the power consumption weights of different maximum points and the previous maximum point as the annual power consumption continuity of each power supply area.

[0033] It should be noted that, in the embodiments of the present invention, linear normalization or normalization function is used for normalization, and the fitting curve is obtained by least squares method or polynomial fitting. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0034] Obtain the power load curves fitted by the power load of each power supply area at the same time period on different days. Select the two power load curves with the highest and lowest positions in all time periods. Calculate the area difference between the two curves through definite integral and perform negative correlation normalization to obtain the multi-period continuous load degree of each power supply area. The annual continuity of power consumption is gain-adjusted according to the multi-period continuous load of each power supply area to obtain the continuous high load performance of each power supply area.

[0035] It should be noted that the annual continuity of power consumption reflects the continuity of the power consumption state of the power supply area throughout the year. The greater the annual continuity of power consumption, the more continuous the load performance. In one embodiment of the present invention, the sum of the positive integer 1 and the multi-period continuous load degree of each power supply area is obtained as the multi-period continuous load regulation degree. The product between the multi-period continuous load regulation degree and the annual continuity of power consumption is calculated and normalized as the continuous high load performance degree of each power supply area.

[0036] It should be noted that, in one embodiment of the present invention, negative correlation normalization can be performed by taking the inverse of the area difference and performing normalization. In other embodiments of the present invention, negative correlation normalization can also be performed through a function. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0037] Preferably, in one embodiment of the present invention, the method for obtaining the load decay performance includes: Obtain the average power load value of each power supply area in all time periods of each day as the average power load level; obtain the minimum point in the curve composed of the average power load levels of all days of the year as the load decay point; According to the average power load level distribution of each load decay point on the left and right sides, the decay node performance of each load decay point is obtained; It should be noted that, in one embodiment of the present invention, the method for obtaining the attenuation node performance degree can be found in Figure 3 , which shows a flow chart of a method for obtaining attenuated node expressiveness, including: Step S301: The average power load level corresponding to the load decay point is taken as the target load, and the first decay degree is obtained according to the first difference between the maximum value of the average power load level in all days and the target load, and the second difference between the average values ​​of all average power load levels on the left and right sides of each load decay point. Both the first difference and the second difference are positively correlated with the first decay degree.

[0038] It should be noted that the first difference between the maximum value of the average electricity load level on all days and the target load reflects the change in peak and valley loads. The greater the change, the greater the first difference, and the more attention should be paid to the load change; the second difference between the average values ​​of all average electricity load levels on the left and right sides of each load decay point reflects the load decay. The larger the second difference, the greater the decay, that is, the greater the first decay degree, and both the first difference and the second difference are positively correlated with the first decay degree.

[0039] It should be noted that in other embodiments of the present invention, the difference between the data can also be reflected by analyzing the ratio. The larger the ratio, the larger the difference between the previous data and the next data, and the greater the difference. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0040] In one embodiment of the present invention, the product of the first difference and the second difference is obtained and normalized and mapped to form the first attenuation degree. Therefore, based on the above basic mathematical operations, a correlation is established between the first difference, the second difference, and the first attenuation degree. That is, the larger the first difference, the greater the attention paid to load changes, and the larger the second difference, the greater the first attenuation degree.

[0041] Step S302: Obtain a range of consecutive days on the left side during which the average power load level is greater than or equal to the target load as a first time interval; obtain a range of consecutive days on the right side during which the average power load level is less than or equal to the target load as a second time interval; obtain a second decay degree based on the first difference and the interval difference between the second time interval and the first time interval, wherein the first difference is negatively correlated with the second decay degree, and the interval difference is positively correlated with the second decay degree; It should be noted that the smaller the first difference between the maximum value of the average power load level and the target load, the less obvious the load data change trend, the more attention is paid to the temporal decay trend change, and the greater the credibility of the interval difference; the interval difference between the second time interval and the first time interval reflects the impact of the load transition time trend. The larger the interval difference, the larger the second time interval, the smaller the time corresponding to the load attenuation point, the longer the subsequent attenuation, and the greater the second attenuation degree. The first difference is negatively correlated with the second attenuation degree, and the interval difference is positively correlated with the second attenuation degree.

[0042] In one embodiment of the present invention, the difference between the positive integer 1 and the first difference is obtained, and the product between the difference result and the interval difference is calculated as the second attenuation degree; based on the above basic mathematical operations, a correlation between the first difference, the interval difference and the second attenuation degree is constructed, that is, the smaller the first difference, the less attention is paid to load changes, the greater the attention is paid to temporal decay, and the larger the interval difference, the greater the second attenuation degree.

[0043] Step S303: Obtain the sum of the first attenuation degree and the second attenuation degree of each load attenuation point as the attenuation node representation degree of each load attenuation point.

[0044] The attenuation node with the largest performance degree among all load attenuation points is selected, and the attenuation node performance degree of the corresponding load attenuation point is used as the load attenuation performance degree of each power supply area.

[0045] Step S3: Based on the changes in daily power consumption in the power supply area on all days of different years, traditional industrial areas and new load areas are screened out; the number of times different power supply areas support and the number of times they are supported are obtained, and combined with the overall expansion sustainability performance of each new load area, the comprehensive expansion dynamic impact of each new load area is obtained.

[0046] Traditional industrial areas usually maintain relatively stable electricity demand over this long period, and the power load in the overall power system changes little, maintaining a relatively stable load state.

[0047] With the development of cities, the power consumption of traditional industrial areas may decline, and new load areas are emerging with greater power consumption. Therefore, traditional industrial areas and new load areas are screened out based on the changes in daily power consumption in power supply areas on all days of different years.

[0048] Preferably, in one embodiment of the present invention, the method for obtaining the traditional industrial area and the new load area includes: Obtain the sum of the daily power consumption of each power supply area on all days of each year as the overall power consumption of each power supply area in each year; Calculate the difference in overall power consumption between the latest historical year and the previous year for each power supply area as the power consumption decline change degree of each power supply area; If the power consumption decay variation degree of the power supply area is less than or equal to the preset decay threshold, the corresponding power supply area is regarded as a traditional industrial area; otherwise, the corresponding power supply area is regarded as a new load area.

[0049] It should be noted that, in one embodiment of the present invention, the preset decay threshold is 0.65; in other embodiments of the present invention, the preset decay threshold may be set according to specific circumstances, which is not limited or elaborated herein.

[0050] In the case of expansion of new load areas, when the power supply area may be overloaded, the spare capacity of the adjacent power supply areas shall be called upon first to meet the power demand, and the dispatch event logs shall be obtained based on the urban power dispatching and management system, which shall be reflected as the number of support or supported times of the area; the overall expansion continuity performance of each new load area shall reflect the continuous growth of the power load in the area, and the greater the continuous growth, the more expansive it is; the comprehensive expansion dynamic impact shall be quantified through the number of support and supported times of different power supply areas and the overall expansion continuity performance of each new load area, reflecting the degree of influence of load expansion on the stability of the power system; the number of support and supported times of different power supply areas shall be obtained, and combined with the overall expansion continuity performance of each new load area, the comprehensive expansion dynamic impact of each new load area shall be obtained.

[0051] Preferably, in one embodiment of the present invention, the method for obtaining the comprehensive expansion dynamic influence includes: For each power supply area, the ratio between the number of support times and the number of supported times is obtained as the power contribution; According to the number of times the adjacent areas of each new load area support the new load area, the number of times the new load area is supported, the power contribution and the overall expansion sustainability performance, the comprehensive expansion dynamic influence of each new load area is obtained. The number of times the adjacent areas support the new load area and the power contribution are negatively correlated with the comprehensive expansion dynamic influence, while the number of times the new load area is supported and the overall expansion sustainability performance are positively correlated with the comprehensive expansion dynamic influence.

[0052] It should be noted that the number of times the adjacent areas of each new load area support the new load area reflects the number of times the new load area is supported by other neighboring power supply areas. The closer to the number of supported times of the new load area, the greater the number of supports for the new load area in the adjacent areas relative to the number of supported times. When power is tight, the possibility of close power supply support is greater, and the expansion impact is smaller; the power contribution reflects the support ability of the new load area to the balance of power supply and demand. The greater the number of supports, the greater the power contribution, the relatively abundant power in the new load area, and the smaller the expansion impact; the overall expansion sustainability reflects the sustainability of load expansion. The greater the expansion sustainability, the greater the expansion impact; therefore, the number of supports and power contributions to the new load area are negatively correlated with the comprehensive expansion dynamic impact, and the number of supports and overall expansion sustainability of the new load area are positively correlated with the comprehensive expansion dynamic impact.

[0053] In one embodiment of the present invention, the ratio of the number of times a new load area is supported by its neighboring regions to the number of times the new load area is supported is obtained as the feasibility of close-range power supply support; the product of the power contribution and the close-range power supply support feasibility is obtained and normalized as the contribution-close-range support linkage coefficient; the difference between the positive integer 1 and the contribution-close-range support linkage coefficient is obtained, and the product of the difference result and the expansion sustainability table limit is calculated as the comprehensive expansion dynamic influence of each new load area. Therefore, based on the above basic mathematical operations, a correlation is established between the number of times a new load area is supported, the power contribution, the number of times a new load area is supported, the overall expansion sustainability performance, and the comprehensive expansion dynamic influence. That is, the smaller the number of times a new load area is supported, the smaller the power contribution, the greater the number of times a new load area is supported, and the greater the overall expansion sustainability performance, the greater the comprehensive expansion dynamic influence obtained.

[0054] Step S4: For the latest historical year, the power saturation capacity impact coefficient is obtained based on the sustained high load performance of different power supply areas, the load decline performance of all traditional industrial areas, and the comprehensive expansion dynamic impact of all new load areas; based on the power saturation capacity impact coefficient and the distribution of power load in different years, the optimized saturation capacity is obtained.

[0055] Considering that the impact of new load areas on power saturation capacity is increasing, while the impact of traditional industrial areas is decreasing, the impact of sustained high-load power demand on power saturation capacity needs to be paid more attention, with the sustained high-load performance as the weight. Influenced by various factors, traditional industrial areas exhibit decline, leading to significant fluctuations in power load, while new load areas experience significant growth, exhibiting expansionary load. Therefore, by analyzing the sustained high-load performance of different power supply areas, the load decline performance of all traditional industrial areas, and the combined expansion dynamic impact of all new load areas, the power saturation capacity impact coefficient is obtained based on the sustained high-load performance of different power supply areas, the load decline performance of all traditional industrial areas, and the combined expansion dynamic impact of all new load areas for the most recent historical year.

[0056] Preferably, in one embodiment of the present invention, the method for obtaining the power saturation capacity influence coefficient includes: Obtaining the product of the sustained high load performance and the load decay performance of all traditional industrial areas as the first performance value; Obtaining a product cumulative value of the sustained high load performance degree and the comprehensive expansion dynamic impact degree of all new load areas as a second performance value; A difference between the first performance value and the second performance value is obtained as a power saturation capacity influence coefficient.

[0057] Saturated capacity reflects the maximum load limit that a region's power system can bear. The distribution of electricity load reveals the growth characteristics of a city's electricity load. Relying solely on historical growth trends may overestimate the actual achievable capacity. The power saturated capacity impact coefficient combines power changes and trend changes to correct the saturated capacity. The optimized saturated capacity is obtained based on the power saturated capacity impact coefficient and the distribution of electricity load in different years.

[0058] Preferably, in one embodiment of the present invention, the method for obtaining the optimized saturation capacity includes: The maximum value of the electricity load in a preset number of historical years is obtained, and the logistic curve is fitted using the nonlinear least squares method to obtain the initial saturation capacity of the latest historical year.

[0059] The initial saturation capacity is gain-adjusted according to the power saturation capacity influence coefficient to obtain the optimized saturation capacity.

[0060] It should be noted that the specific nonlinear least squares method for fitting the logistic curve is a technical means well known to those skilled in the art and will not be described in detail here.

[0061] It should be noted that, in one embodiment of the present invention, a preset number of historical years, including the latest historical year and adjacent historical years, are selected, and the preset number is 5. In other embodiments of the present invention, the preset number can be set according to specific circumstances, and is not limited or elaborated here.

[0062] In one embodiment of the present invention, the sum of a positive integer 1 and a power saturation capacity influence coefficient is obtained as the influence weight; The product between the initial saturation capacity and the influence weight is obtained as the optimized saturation capacity.

[0063] Step S5: forecasting the city's electricity demand based on the Logistic model after optimizing the saturated capacity.

[0064] By optimizing the saturation capacity parameters of the traditional Logistic model, a more realistic urban power demand forecast curve is generated, which improves the effect of power demand forecasting and is conducive to urban development planning.

[0065] In summary, the present invention analyzes daily power consumption and load trends to obtain the overall expansion sustainability, sustained high load, and load decline performance of each power supply area, and screens out traditional industrial areas and new load areas. The number of times each power supply area supports and is supported is obtained, and combined with the overall expansion sustainability of each new load area, the comprehensive expansion dynamic impact of each new load area is obtained. For the most recent historical year, the power saturation capacity impact coefficient is obtained, and combined with the distribution of power load across different historical years, the optimized saturation capacity is obtained to predict urban power demand. By obtaining accurate saturation capacity when using the logistic model for prediction, the present invention improves the accuracy of power demand forecasts.

[0066] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for predicting urban power demand based on a logistic model, characterized in that: The method comprises: The daily power consumption of different power supply areas in the city and the power load in each time period of each day are obtained; based on the daily power consumption distribution of the power supply areas on all days of the year and the power load change trends in the same time period on different days, the overall expansion sustainability performance, sustained high load performance and load decline performance of each power supply area are obtained; based on the daily power consumption changes of the power supply areas on all days of different years, traditional industrial areas and new load areas are screened out; the number of support times and the number of times supported are obtained for different power supply areas, and the comprehensive expansion dynamic impact of each new load area is obtained in combination with the overall expansion sustainability performance; for the latest historical year, the power saturation capacity influence coefficient is obtained based on the sustained high load performance of different power supply areas, the load decline performance of all traditional industrial areas, and the comprehensive expansion dynamic impact of all new load areas; based on the power saturation capacity influence coefficient and the distribution of power load in different years, the optimized saturation capacity is obtained; and the urban power demand is predicted based on the logistic model after the optimized saturation capacity.

2. The method for predicting urban power demand based on the Logistic model according to claim 1, characterized in that: The method for obtaining the overall expansion continuous performance includes: In the neighborhood of the latest historical year, the accumulated daily power consumption of each power supply area on all days of each month is obtained and normalized to serve as the power consumption weight of each power supply area in each month. Obtain the difference in power consumption weights between different months and the previous month as the weight difference; consider all months with consecutive positive weight differences to constitute an expansion development cycle; calculate the mean of the power consumption weights of all months in each expansion development cycle as the expansion power consumption weight degree; calculate the mean of the ratio of the cumulative daily power consumption between different months and the previous month in each expansion development cycle as the expansion coefficient; The expansion power consumption weight is gain-adjusted according to the expansion coefficient of each expansion development cycle as the local expansion sustainability performance of each expansion development cycle; the local expansion sustainability performance with the largest value in all expansion development cycles is selected as the overall expansion sustainability performance of each power supply area.

3. The method for predicting urban power demand based on a logistic model according to claim 1, characterized in that: The method for obtaining the sustained high-load performance includes: Based on the daily power consumption distribution of each power supply area on all days of the year, the annual power consumption continuity of each power supply area is obtained; Obtain the power load curves fitted by the power load of each power supply area at the same time period on different days. Select the two power load curves with the highest and lowest positions in all time periods. Calculate the area difference between the two curves through definite integral and perform negative correlation normalization to obtain the multi-period continuous load degree of each power supply area. The annual power consumption continuity is gain-adjusted based on the multi-period continuous load of each power supply area to obtain the continuous high load performance of each power supply area.

4. The method for predicting urban power demand based on a logistic model according to claim 3, characterized in that: The method for obtaining the annual duration of power consumption includes: Normalize the daily power consumption of each power supply area to serve as the daily power consumption weight of each power supply area; Obtain a fitting curve for the power consumption weights of each power supply area on all days, obtain the maximum point of the fitting curve, and obtain the cumulative value of the difference between the power consumption weights of different maximum points and the previous maximum point as the annual power consumption continuity of each power supply area.

5. The method for predicting urban power demand based on a logistic model according to claim 1, characterized in that: The method for obtaining the load decay performance includes: Obtain the average power load value of each power supply area in all time periods of each day as the average power load level; obtain the minimum point in the curve composed of the average power load levels of all days of the year as the load decay point; According to the average power load level distribution of each load decay point on the left and right sides, the decay node performance of each load decay point is obtained; The attenuation node with the largest performance degree among all load attenuation points is selected, and the attenuation node performance degree of the corresponding load attenuation point is used as the load attenuation performance degree of each power supply area.

6. The method for predicting urban power demand based on the Logistic model according to claim 5, characterized in that: The method for obtaining the attenuation node expression degree includes: The average power load level corresponding to the load decay point is taken as the target load. The first decay degree is obtained according to the first difference between the maximum value of the average power load level in all days and the target load, and the second difference between the average values ​​of all the average power load levels on the left and right sides of each load decay point. Both the first difference and the second difference are positively correlated with the first decay degree. Obtain the range of consecutive days on the left side during which the average power load level is greater than or equal to the target load as the first time interval; obtain the range of consecutive days on the right side during which the average power load level is less than or equal to the target load as the second time interval; obtain the second decay degree based on the first difference and the interval difference between the second time interval and the first time interval, where the first difference is negatively correlated with the second decay degree, and the interval difference is positively correlated with the second decay degree; The sum of the first attenuation degree and the second attenuation degree of each load reduction point is obtained as the attenuation node representation degree of each load reduction point.

7. The method for predicting urban power demand based on a logistic model according to claim 1, characterized in that: The method for obtaining the traditional industrial area and the new load area includes: Obtain the sum of the daily power consumption of each power supply area on all days of each year as the overall power consumption of each power supply area in each year; Calculate the difference in overall power consumption between the latest historical year and the previous year for each power supply area as the power consumption decline change degree of each power supply area; If the power consumption decay variation degree of the power supply area is less than or equal to the preset decay threshold, the corresponding power supply area is regarded as a traditional industrial area; otherwise, the corresponding power supply area is regarded as a new load area.

8. The method for predicting urban power demand based on a logistic model according to claim 1, characterized in that: The method for obtaining the comprehensive expansion dynamic influence degree includes: For each power supply area, the ratio between the number of support times and the number of supported times is obtained as the power contribution; According to the number of times the adjacent areas of each new load area support the new load area, the number of times the new load area is supported, the power contribution and the overall expansion sustainability performance, the comprehensive expansion dynamic influence of each new load area is obtained. The number of times the adjacent areas support the new load area and the power contribution are negatively correlated with the comprehensive expansion dynamic influence, while the number of times the new load area is supported and the overall expansion sustainability performance are positively correlated with the comprehensive expansion dynamic influence.

9. The method for predicting urban power demand based on a logistic model according to claim 1, characterized in that: The method for obtaining the power saturation capacity influence coefficient includes: Obtaining the product of the sustained high load performance and the load decay performance of all traditional industrial areas as the first performance value; Obtaining a product cumulative value of the sustained high load performance degree and the comprehensive expansion dynamic impact degree of all new load areas as a second performance value; A difference between the first performance value and the second performance value is obtained as a power saturation capacity influence coefficient.

10. The method for predicting urban power demand based on a Logistic model according to claim 1, characterized in that: The method for obtaining the optimized saturation capacity includes: Obtain the maximum value of the electricity load in a preset number of historical years, use the nonlinear least squares method to fit the logistic curve, and obtain the initial saturation capacity of the latest historical year; The initial saturation capacity is gain-adjusted according to the power saturation capacity influence coefficient to obtain the optimized saturation capacity.

Images