Village and town saturated annual load prediction method and system

Abstract

The invention discloses a township saturated annual load prediction method. The method comprises the following steps: acquiring power consumption data of township users in the current year; screening the electricity consumption data of the town users to obtain a sample data set and a non-sample data set; based on the average value and the standard deviation of the sample data set, classifying the sample data, and performing saturated year data prediction according to the classification condition to obtain a predicted sample data set; constructing a town saturated annual load prediction model; and in combination with the non-sample data set and the prediction sample data set, through the town saturated annual load prediction model, predicting the town saturated annual load to obtain a prediction result. The invention also discloses a system for realizing the method for predicting the saturated annual load of the village and town. The method provided by the invention solves the influence of prediction uncertainty and variability caused by the development of village and town regions, provides reference for load prediction in a saturation year, and improves the power supply capability and stability of a novel urbanization construction power distribution network.

Description

Technical Field

[0001] The invention belongs to the technical field of composite forecasting of electric power market, and in particular relates to a method and system for forecasting load in a township saturated year. Background Art

[0002] Saturated annual load refers to the scale of a city or region's power load when it reaches saturation, that is, the load level when the power demand reaches the maximum capacity. This state reflects the balance point between the power demand and supply of a city or region, and is an important concept in the power system. The prediction of saturated annual load is of great significance for the long-term planning of urban power grid planning, power supply configuration, transmission network and distribution network. It can help planners avoid duplication of construction and waste of resources, and ensure the stable operation of the power system and the rational use of resources.

[0003] With the continuous development and construction of cities, the prediction of saturated load of urban power grids is becoming increasingly important. Saturated load plays an indispensable role in guiding urban network planning and substation layout. At present, with the advancement of new urbanization construction, higher requirements are put forward for the power supply capacity and reliability of distribution networks, especially urban distribution networks. Realizing reliable power supply, high-quality services, and improving the electrification level of residents have become the responsibility of power supply in the new era. For township areas, due to the uncertainty and variability in their development process, there is currently no definite method to predict their saturated load. Therefore, a method and system for predicting saturated annual load for rural township areas is needed. Summary of the invention

[0004] One of the purposes of the present invention is to provide a method for predicting the saturated annual load in towns and villages, which can solve the influence of the uncertainty and variability of the saturated load in the development process of towns and villages, and improve the stability of the power grid.

[0005] The second object of the present invention is to provide a system for realizing the method for predicting the saturated annual load of townships.

[0006] The present invention provides a method for predicting the saturated annual load of a township, comprising the following steps:

[0007] S1. Obtain the electricity consumption data of township users in the current year;

[0008] S2. Filter the electricity consumption data of township users obtained in step S1 to obtain a sample data set and a non-sample data set;

[0009] S3. Based on the mean and standard deviation of the sample data set obtained in step S2, the sample data is classified, and then the saturated year data is predicted according to the classification to obtain a predicted sample data set;

[0010] S4. Construct a saturated annual load forecasting model for townships;

[0011] S5. Combining the non-sample data set obtained in step S2 and the predicted sample data set obtained in step S3, the township saturated annual load prediction model obtained in step S4 is used to predict the township saturated annual load and obtain a prediction result.

[0012] Step S2 is specifically as follows: according to the electricity consumption data of township users obtained in step S1, when the user's annual electricity consumption is within the preset range and the maximum load utilization hours are less than the preset threshold, the user is judged as sample data, otherwise it is non-sample data, and finally a sample data set and a non-sample data set are obtained.

[0013] Step S3 includes the following steps:

[0014] Calculate the mean and standard deviation of the sample data set;

[0015] According to the mean value a and standard deviation σ of the sample data set, the sample data in the sample data set are divided into four grades from low to high according to the annual electricity consumption, and the annual electricity consumption of the sample is divided into four intervals: 0~a-σ, a-σ~a, a~a+σ and greater than a+σ, and the load index corresponding to the four grades is calculated; for the load index of the same interval, the average value of the maximum active power of all users on New Year's Eve in the interval is used as the load index corresponding to the interval;

[0016] When predicting to the saturation year, the user data of the iiith level in the sample data set will jump to the iii+1th level, and the load index of the iii+1th level will be used for subsequent calculations. The highest level remains unchanged to obtain the predicted sample data set; the electricity consumption of the non-sample data set will not increase with the growth of time. The jump of the electricity load index of the non-sample data set is not considered, and the current existing load index is used for subsequent calculations.

[0017] The objective function of the township saturated annual load forecasting model in step S4 is expressed by the following formula:

[0018] P Tmax =∑(Y 1max *T1+...+Y imax *T i )

[0019] P Lmax =∑(P T1max *T L1 +...+P Tiimax *T Lii )

[0020] P Amax =∑(P L1max *T A1 +...+P Ljmax *T Aj )

[0021] Among them, P Tmax Y is the saturated annual load of a single substation in a township; 1max is the maximum load of user 1 in a certain area of ​​a township; T1 is the simultaneous rate of user 1 in a certain area of ​​a township; Y imax is the maximum load of user i in a township area; T i is the simultaneous rate of user i in a certain area of ​​a township; P Lmax P is the saturated annual load of a single line in a township; T1max is the saturated annual load of the township area; T L1 is the simultaneous rate of township stations 1; P Tiimax is the saturated annual load of township area ii; T Lii is the simultaneous rate of township area ii; P Amax P is the maximum load of the township in saturated year; L1max T is the maximum annual load of township line 1; A1 is the simultaneous rate of township line 1; P Ljmax T is the maximum load of the township line j in saturation year; Aj is the simultaneity rate of township line j;

[0022] First, calculate the simultaneous rate of user i in a certain area in a township using the following formula:

[0023]

[0024] Among them, P irmax is the maximum load of user i in the whole year; P ir is the load of user i when the maximum load of the area to which user i belongs occurs.

[0025] According to the obtained simultaneous rate of user i in a certain substation in a township, combined with the maximum load of users in the non-sample data set and the predicted sample data set, the maximum load of each substation in the township in the saturated year is calculated through the objective function of the township saturated year load prediction model;

[0026] Calculate the simultaneous rate of township stations ii in the township using the following formula:

[0027]

[0028] Among them, P rLiimax is the maximum load of area ii in the whole year; P rLii is the load of area ii when the maximum load of the line belonging to area ii occurs;

[0029] According to the obtained simultaneous rate of township area ii, combined with the maximum load of each area in the saturated year obtained in the previous step, the maximum load of each line in the township in the saturated year is calculated through the objective function of the township saturated year load forecasting model;

[0030] Calculate the simultaneous rate of township line j using the following formula:

[0031]

[0032] Among them, P rAjmax is the maximum load of line j in the whole year; P rAj is the load of line j when the maximum load of the township occurs;

[0033] According to the obtained synchronization rate of township line j, combined with the maximum load of each line in the saturated year obtained in the previous step, the maximum load of the township in the saturated year is calculated through the objective function of the township saturated year load forecasting model.

[0034] The present invention also provides a system for implementing the township saturated annual load forecasting method, comprising a power consumption data acquisition module, a power consumption data screening module, a sample data classification forecasting module, a model building module, and a township saturated annual load forecasting module connected in series in sequence;

[0035] The electricity consumption data collection module obtains the electricity consumption data of township users in the current year and uploads the data to the electricity consumption data screening module;

[0036] The electricity consumption data screening module screens the electricity consumption data of township users according to the received data, obtains sample data sets and non-sample data sets, and uploads the data to the sample data classification prediction module;

[0037] The sample data classification prediction module classifies the sample data based on the received data, the mean value and standard deviation of the sample data set, and then predicts the saturated year data according to the classification situation to obtain the predicted sample data set, and uploads the data to the model construction module;

[0038] The model building module builds a township saturated annual load forecasting model based on the received data, and uploads the data to the township saturated annual load forecasting module;

[0039] The township saturated annual load forecasting module predicts the township saturated annual load based on the received data, combined with the non-sample data set and the predicted sample data set, through the township saturated annual load forecasting model, and obtains the forecast result.

[0040] The present invention discloses a method and system for predicting load in a saturated year in a township, which solves the impact of prediction uncertainty and variability caused by development in township areas, provides a reference for load prediction in a saturated year, and improves the power supply capacity and stability of the distribution network for new urbanization construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the process of the present invention;

[0042] Figure 2 It is a schematic diagram of the structure of the system of the present invention. DETAILED DESCRIPTION

[0043] The present invention provides a method for predicting the saturated annual load of a township, and its flow chart is as follows: Figure 1 As shown, the following steps are included:

[0044] S1. Obtain the electricity consumption data of township users in the current year;

[0045] S2. Filter the electricity consumption data of township users obtained in step S1 to obtain sample data sets and non-sample data sets, specifically:

[0046] According to the electricity consumption data of township users obtained in step S1, when the user's annual electricity consumption is within the preset range and the maximum load utilization hours are less than the preset threshold, the user is determined to be sample data, otherwise it is non-sample data, and finally a sample data set and a non-sample data set are obtained.

[0047] S3. Based on the mean value and standard deviation of the sample data set obtained in step S2, the sample data is classified, and then the saturated year data is predicted according to the classification to obtain a predicted sample data set, including the following steps:

[0048] Calculate the mean and standard deviation of the sample data set;

[0049] According to the mean value a and standard deviation σ of the sample data set, the sample data in the sample data set are divided into four levels from low to high according to the annual electricity consumption, and the annual electricity consumption of the sample is divided into four intervals: 0~a-σ, a-σ~a, a~a+σ and greater than a+σ, and the load indicators corresponding to the four levels are calculated;

[0050] Because the annual electricity consumption of rural users meets the characteristics of normal distribution, the annual electricity consumption of the sample is divided into four intervals: 0~a-σ, a-σ~a, a~a+σ and greater than a+σ. According to the probability distribution law of normal distribution, it can be seen that the proportion of the first interval is 16%, the second interval is 34%, the third interval is 34%, and the fourth interval is 16%; for the load index of the same interval, the average value of the maximum active power of all users in the interval on New Year's Eve is used as the load index corresponding to the interval.

[0051] When predicting to the saturation year, the user data of the iiith level in the sample data set will jump to the iii+1th level, and the load index of the iii+1th level will be used for subsequent calculations. The highest level remains unchanged to obtain the predicted sample data set; the electricity consumption of the non-sample data set will not increase with the growth of time. The jump of the electricity load index of the non-sample data set is not considered, and the current existing load index is used for subsequent calculations.

[0052] S4. Construct a saturated annual load forecasting model for townships;

[0053] The objective function of the township saturated year load forecasting model is expressed using the following formula:

[0054] P Tmax =∑(Y 1max *T1+...+Y imax *T i )

[0055] P Lmax =∑(P T1max *T L1 +...+P Tiimax *T Lii )

[0056] P Amax =∑(P L1max *T A1 +...+P Ljmax *T Aj )

[0057] Among them, P Tmax Y is the saturated annual load of a single substation in a township; 1max is the maximum load of user 1 in a certain area of ​​a township; T1 is the simultaneous rate of user 1 in a certain area of ​​a township; Y imax is the maximum load of user i in a township area; T i is the simultaneous rate of user i in a certain area of ​​a township; P Lmax P is the saturated annual load of a single line in a township; T1max is the saturated annual load of the township area; T L1 is the simultaneous rate of township stations 1; P Tiimax is the saturated annual load of township area ii; T Lii is the simultaneous rate of township area ii; P Amax P is the maximum load of the township in saturated year; L1max T is the maximum annual load of township line 1; A1 is the simultaneous rate of township line 1; P Ljmax T is the maximum load of the township line j in saturation year; Aj is the simultaneity rate of township line j;

[0058] First, calculate the simultaneous rate of user i in a certain area in a township using the following formula:

[0059]

[0060] Among them, P irmax is the maximum load of user i in the whole year; P ir is the load of user i when the maximum load of the area to which user i belongs occurs.

[0061] According to the obtained simultaneous rate of user i in a certain substation in a township, combined with the maximum load of users in the non-sample data set and the predicted sample data set, the maximum load of each substation in the township in the saturated year is calculated through the objective function of the township saturated year load prediction model;

[0062] Calculate the simultaneous rate of township stations ii in the township using the following formula:

[0063]

[0064] Among them, P rLiimax is the maximum load of area ii in the whole year; P rLii is the load of area ii when the maximum load of the line belonging to area ii occurs;

[0065] According to the obtained simultaneous rate of township area ii, combined with the maximum load of each area in the saturated year obtained in the previous step, the maximum load of each line in the township in the saturated year is calculated through the objective function of the township saturated year load forecasting model;

[0066] Calculate the simultaneous rate of township line j using the following formula:

[0067]

[0068] Among them, P rAjmax is the maximum load of line j in the whole year; P rAj is the load of line j when the maximum load of the township occurs;

[0069] According to the obtained synchronization rate of township line j, combined with the maximum load of each line in the saturated year obtained in the previous step, the maximum load of the township in the saturated year is calculated through the objective function of the township saturated year load forecasting model.

[0070] S5. Combining the non-sample data set obtained in step S2 and the predicted sample data set obtained in step S3, the township saturated annual load prediction model obtained in step S4 is used to predict the township saturated annual load and obtain a prediction result.

[0071] The present invention also provides a system for realizing the method for predicting the saturated annual load of a township, and its structural schematic diagram is shown in FIG. Figure 2As shown, it includes a power consumption data collection module, a power consumption data screening module, a sample data classification prediction module, a model building module, and a township saturated annual load prediction module which are connected in series in sequence;

[0072] The electricity consumption data collection module obtains the electricity consumption data of township users in the current year and uploads the data to the electricity consumption data screening module;

[0073] The electricity consumption data screening module screens the electricity consumption data of township users according to the received data, obtains sample data sets and non-sample data sets, and uploads the data to the sample data classification prediction module;

[0074] The sample data classification prediction module classifies the sample data based on the received data, the mean value and standard deviation of the sample data set, and then predicts the saturated year data according to the classification situation to obtain the predicted sample data set, and uploads the data to the model construction module;

[0075] The model building module builds a township saturated annual load forecasting model based on the received data, and uploads the data to the township saturated annual load forecasting module;

[0076] The township saturated annual load forecasting module predicts the township saturated annual load based on the received data, combined with the non-sample data set and the predicted sample data set, through the township saturated annual load forecasting model, and obtains the forecast result.

[0077] The method of the present invention is further described below in conjunction with an embodiment:

[0078] A power supply station has 207 stations in total, supplying 19,378 users. In real life, a large amount of continuous data such as measured values ​​often presents a normal distribution shape, so the annual electricity consumption of rural users also meets the characteristics of normal distribution. After screening, the 19,378 households can be divided into two categories: the first category is sample data, which is 10,598 households; the second category is non-sample data, which is 8,780 households, including 2,776 non-residential users and 6,004 residential users.

[0079] For the sample data, the sample mean value a and standard deviation σ are calculated, and based on this, the load of the sample data is divided into four levels. By the future year, the first-level users of the sample data will jump to the second-level users, the second-level users will jump to the third-level users, the third-level users will jump to the fourth-level users, and the fourth-level users will remain unchanged.

[0080] For non-sample data, it is assumed that the electricity consumption of non-residential users will remain unchanged until the future year. Most of the residential users in the non-sample have an electricity consumption of less than 10 kWh, and a small part has an electricity consumption of more than 30,000 kWh, which is in an extremely low or extremely high state. Therefore, when calculating the load of residential users in the non-sample, the leap in electricity consumption indicators is not taken into account.

[0081] Calculate the load of sample data and non-sample data respectively, and then get the total load of all users in the town in the future year. Calculate the simultaneous rate of each user to attribute the user load to the substation, and finally summarize it by substation. Based on this, the substation is grouped into villages, and the heavy and overload conditions of the substations in the 25 villages (communities) of the town are counted. If there are both lightly loaded and heavily overloaded substations in a village, consider rotating the substations. The specific steps are as follows:

[0082] In order to ensure that the data samples studied are rural residents' electricity consumption, the original data samples need to be screened. After investigating and analyzing the electricity consumption of residents, the following two screening indicators were formulated. After screening, the number of households in the town is 10,598: annual electricity consumption must be greater than 10 kWh and less than 30,000 kWh; the maximum load utilization hours must be less than 3,000 hours.

[0083] After quantitative analysis of the sample data after screening, the average value a of the sample power data is 2044 kWh, the standard deviation σ is 1648 kWh, and the annual power of the sample is divided into four intervals: 0~a-σ, a-σ~a, a~a+σ and greater than a+σ, and the corresponding four-grade load indicators are obtained. The simultaneous rate of the substation is 41.36%. The data after grading is shown in Table 1.

[0084] Table 1 Classification results of township sample data

[0085]

[0086]

[0087] Taking into account the load indicators of the township and the existing four-tier load classification indicators, the sample data of residents in the prospect year shows that the first-tier users will jump to the second-tier users, the second-tier users will jump to the third-tier users, the third-tier users will jump to the fourth-tier users, and the fourth-tier users will remain unchanged. The predicted sample data is shown in Table 2.

[0088] Table 2 Township prediction sample data

[0089]

[0090] There are 8,780 households in the non-sample data, including 2,776 non-residential users and 6,004 residential users. Assuming that the electricity consumption of non-residential users remains unchanged in the future year, the current maximum annual load of non-residential users after taking into account the simultaneity rate is 1,300 kilowatts.

[0091] Most of the residential users in the non-sample consume less than 10 kWh of electricity, and a small part consumes more than 30,000 kWh, which is in an extremely low or extremely high state. Therefore, when calculating the load of residential users in the non-sample, the jump in electricity consumption indicators is not considered, and the corresponding four-level load indicators are calculated based on the sample data analysis. The non-sample data is shown in Table 3.

[0092] Table 3 Classification results of township non-sample data

[0093]

[0094] According to the method of the present invention, it is calculated that by the prospective year, the maximum load of the township will be 28,000 kilowatts.

Claims

1. A method for predicting the saturated annual load of a township, characterized in that: The following steps are involved: S1. Obtain the electricity consumption data of township users in the current year; S2. Filter the electricity consumption data of township users obtained in step S1 to obtain a sample data set and a non-sample data set; S3. Based on the mean and standard deviation of the sample data set obtained in step S2, the sample data is classified, and then the saturated year data is predicted according to the classification to obtain a predicted sample data set; S4. Construct a saturated annual load forecasting model for townships; S5. Combining the non-sample data set obtained in step S2 and the predicted sample data set obtained in step S3, the township saturated annual load prediction model obtained in step S4 is used to predict the township saturated annual load and obtain a prediction result.

2. The method for predicting the saturated annual load of a township according to claim 1, characterized in that: Step S2 is specifically as follows: according to the electricity consumption data of township users obtained in step S1, when the user's annual electricity consumption is within the preset range and the maximum load utilization hours are less than the preset threshold, the user is judged as sample data, otherwise it is non-sample data, and finally a sample data set and a non-sample data set are obtained.

3. The method for predicting the saturated annual load of a township according to claim 1, characterized in that: Step S3 includes the following steps: Calculate the mean and standard deviation of the sample data set; According to the mean value a and standard deviation σ of the sample data set, the sample data in the sample data set are divided into four grades from low to high according to the annual electricity consumption, and the annual electricity consumption of the sample is divided into four intervals: 0~a-σ, a-σ~a, a~a+σ and greater than a+σ, and the load index corresponding to the four grades is calculated; for the load index of the same interval, the average value of the maximum active power of all users on New Year's Eve in the interval is used as the load index corresponding to the interval; When predicting to the saturation year, the user data of the iiith level in the sample data set will jump to the iii+1th level, and the load index of the iii+1th level will be used for subsequent calculations. The highest level remains unchanged to obtain the predicted sample data set; the electricity consumption of the non-sample data set will not increase with the growth of time. The jump of the electricity load index of the non-sample data set is not considered, and the current existing load index is used for subsequent calculations.

4. The method for predicting the saturated annual load of a township according to claim 1, characterized in that: The objective function of the township saturated annual load forecasting model in step S4 is expressed by the following formula: P Tmax =∑(Y 1max *T1+...+Y imax *T i ) P Lmax =∑(P T1max *T L1 +...+P Tiimax *T Lii ) P Amax =∑(P L1max *T A1 +...+P Ljmax *T Aj ) Among them, P Tmax Y is the saturated annual load of a single substation in a township; 1max is the maximum load of user 1 in a certain area of ​​a township; T1 is the simultaneous rate of user 1 in a certain area of ​​a township; Y imax is the maximum load of user i in a township area; T i is the simultaneous rate of user i in a certain area of ​​a township; P Lmax P is the saturated annual load of a single line in a township; T1max is the saturated annual load of the township area; T L1 is the simultaneous rate of township stations 1; P Tiimax is the saturated annual load of township area ii; T Lii is the simultaneous rate of township area ii; P Amax P is the maximum load of the township in saturated year; L1max T is the maximum annual load of township line 1; A1 is the simultaneous rate of township line 1; P Ljmax T is the maximum load of the township line j in saturation year; Aj is the simultaneity rate of township line j.

5. The method for predicting the saturated annual load of a township according to claim 4, characterized in that: Calculate the simultaneous rate of user i in a certain area of ​​a township using the following formula: Among them, P irmax is the maximum load of user i in the whole year; P ir is the load of user i when the maximum load of the area to which user i belongs occurs.

6. The method for predicting the saturated annual load of a township according to claim 4, characterized in that: According to the obtained simultaneous rate of user i in a certain substation in a township, combined with the maximum load of users in the non-sample data set and the predicted sample data set, the maximum load of each substation in the township in the saturated year is calculated through the objective function of the township saturated year load prediction model; Calculate the simultaneous rate of township area ii using the following formula: Among them, P rLiimax is the maximum load of area ii in the whole year; P rLii It is the load of substation ii when the maximum load of the line belonging to substation ii occurs.

7. The method for predicting the saturated annual load of a township according to claim 4, characterized in that: According to the obtained simultaneous rate of township area ii, combined with the maximum load of each area in the saturated year obtained in the previous step, the maximum load of each line in the township in the saturated year is calculated through the objective function of the township saturated year load forecasting model; Calculate the simultaneous rate of township line j using the following formula: Among them, P rAjmax is the maximum load of line j in the whole year; P rAj is the load of line j when the maximum load of the township occurs; According to the obtained synchronization rate of each line, combined with the maximum load of each line in the saturated year obtained in the previous step, the maximum load of the township in the saturated year is calculated through the objective function of the township saturated year load forecasting model.

8. A system for implementing the method for predicting the saturated annual load of a township as claimed in any one of claims 1 to 7, characterized in that: It includes a power consumption data collection module, a power consumption data screening module, a sample data classification prediction module, a model building module, and a township saturated annual load prediction module which are connected in series in sequence; The electricity consumption data collection module obtains the electricity consumption data of township users in the current year and uploads the data to the electricity consumption data screening module; The electricity consumption data screening module screens the electricity consumption data of township users according to the received data, obtains sample data sets and non-sample data sets, and uploads the data to the sample data classification prediction module; The sample data classification prediction module classifies the sample data based on the received data, the mean value and standard deviation of the sample data set, and then predicts the saturated year data according to the classification situation to obtain the predicted sample data set, and uploads the data to the model construction module; The model building module builds a township saturated annual load forecasting model based on the received data, and uploads the data to the township saturated annual load forecasting module; The township saturated annual load forecasting module predicts the township saturated annual load based on the received data, combined with the non-sample data set and the predicted sample data set, through the township saturated annual load forecasting model, and obtains the forecast result.

Images