An ordered electricity consumption peak-shaving scheme optimization method considering enterprise production characteristics and multi-time scales

Abstract

The application provides an ordered power consumption peak-shaving scheme optimization method considering enterprise production characteristics and multi-time scales, which comprises the following steps: determining indexes for evaluating enterprise power consumption value according to load gaps under different ordered power consumption peak-shaving scheme levels in a future week and basic information of ordered power consumption of the enterprise, calculating enterprise power consumption value coefficients and unit power production values according to the index data, grouping and clustering all enterprises participating in ordered power consumption according to power consumption attributes and production modes and historical daily load data, respectively, constructing an ordered power consumption peak-shaving scheme optimization model considering enterprise production characteristics and multi-time scales according to the enterprise power consumption value coefficients, the unit power production values, predicted load data and clustering results, and solving the optimization model by using a branch and bound method to obtain ordered power consumption peak-shaving schemes under several early warning levels. The application considers differences in enterprise production characteristics and can realize more refined load response.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of load response, and relates to an orderly power utilization peak avoidance scheme optimization method considering enterprise production characteristics and multiple time scales. BACKGROUND

[0002] Orderly power utilization as a load response means refers to work of maintaining the stable order of power supply and utilization by means of administrative measures, economic means and technical methods to control part of power utilization demand according to law under the conditions of insufficient power supply, emergencies and the like. According to the principle of hierarchical regulation, layer-by-layer inclusion and step-by-step progression, an orderly power utilization peak avoidance scheme is prepared in advance to cope with different power gap grades of early warning and reduce the power supply and demand gap caused power loss. A scientific and reasonable orderly power utilization peak avoidance scheme is crucial to ensure the safe operation of the power grid and the stability of the power supply order.

[0003] The traditional orderly power utilization peak avoidance scheme relies on manpower, has large workload and low efficiency, and the main thought is to group users according to experience, and the differences in production modes and power utilization characteristics of enterprises are not considered. The existing thought for optimizing the orderly power utilization peak avoidance scheme is to score the enterprises according to indexes, sort them according to the scores, accumulate the peak avoidance amount in order, and stop until the power gap value of the early warning grade is reached, but the indexes are relatively single, and the benefit index is mainly used to measure the power utilization value of the user. This makes the orderly power utilization peak avoidance scheme lack of scientificity and flexibility, which is not conducive to the optimal allocation of power resources. SUMMARY

[0004] To solve the deficiencies in the existing method, the application provides an orderly power utilization peak avoidance scheme optimization method considering enterprise production characteristics and multiple time scales. The method comprehensively evaluates the power utilization value of the enterprise from multiple aspects, considers the production mode and power utilization characteristics of the enterprise, is beneficial to improving the operability and scientificity of the scheme, realizes the optimization of the orderly power utilization peak avoidance scheme under the A-F early warning grade, and saves time and labor compared with the artificial scheme. When the orderly power utilization peak avoidance scheme is optimized, the application considers the influence of the enterprise user on the environment in the production process, and the energy consumption and pollution emission are included in the index for evaluating the power utilization value of the enterprise. At the same time, the production mode and production characteristics of the enterprise are comprehensively considered to improve the operability, scientificity and fairness of the orderly power utilization peak avoidance scheme.

[0005] The application realizes the following technical scheme:

[0006] An orderly power utilization peak avoidance scheme optimization method considering enterprise production characteristics and multiple time scales comprises the following steps:

[0007] Step 1, according to the target power grid area future one week different orderly electricity peak avoidance scheme level under the load gap and industrial and commercial enterprises basic information of orderly electricity, determine the index for evaluating enterprise electricity value, according to the index data, calculate enterprise electricity value coefficient and unit electricity production value;

[0008] Step 2, according to the grouping of enterprise electricity attribute, after grouping, further grouping according to the enterprise production mode of the enterprise, and according to the predicted future one week of daily load similarity, using K-means method to cluster the continuous production type enterprise of each group;

[0009] Step 3, according to the grouping and clustering results and enterprise electricity value coefficient and unit electricity production value, establish the orderly electricity peak avoidance scheme optimization model considering enterprise production characteristics and multi time scale and solve, get A-F six early warning level under the orderly electricity peak avoidance scheme.

[0010] In the above technical solution, further, in step 1, the basic information of industrial and commercial enterprises orderly electricity includes historical annual load data, predicted future one week daily load data, enterprise production mode, enterprise electricity attribute, annual main business income, enterprise land area, enterprise real tax, enterprise added value, energy consumption, annual average number of employees, enterprise annual output value, research and development fund expenditure, and pollution discharge right data. The index for evaluating enterprise electricity value includes per mu tax, per mu added value, per mu output value, unit energy consumption added value, full-time labor rate, unit pollution discharge right added value, and research and development fund expenditure proportion of main business income.

[0011] In step 1, the specific calculation formula of enterprise electricity value coefficient is as follows:

[0012]

[0013] Wherein, S i is the electricity value coefficient of the i th enterprise, W g is the contribution of the g th index in user electricity value, P ig is the data of the g th index of the i th enterprise.

[0014] In step 1, the unit electricity production value is calculated by dividing the annual output value by the annual electricity consumption.

[0015] In step 2, according to the enterprise electricity attribute, the enterprise is divided into four groups, each group is a kind, specifically, A class is priority guarantee class, B class is general guarantee class, C class is general restriction class, D class is key restriction class; When calling users in orderly electricity peak avoidance scheme, call in the order of D class, C class, B class and A class;

[0016] In step 3, the construction of the ordered electricity consumption peak-shaving scheme optimization model considering the enterprise production characteristics and multi-time scale includes two stages, specifically including:

[0017] 1) Establishing an optimization model of the ordered electricity consumption peak-shaving scheme under the week time scale; the optimization model of the ordered electricity consumption peak-shaving scheme under the week time scale can determine the ordered electricity consumption peak-shaving scheme of the continuous production type enterprise and the non-continuous production type enterprise according to the size of the future one-week power gap. The way for the non-continuous production enterprise to participate in the ordered electricity consumption is week peak-shaving, and the future one-week shutdown days and specific dates are optimized; the way for the continuous production enterprise to participate in the ordered electricity consumption is day peak-shaving, and the future one-week total load reduction ratio is optimized.

[0018] The objective function of the optimization model of the ordered electricity consumption peak-shaving scheme under the week time scale is to minimize the cost caused by power outage to enterprise users, and the constraint conditions include power balance constraint, corresponding constraints of user number between ordered electricity consumption peak-shaving schemes under different levels (A-F level), continuous production enterprise reduction ratio, and non-continuous production enterprise shutdown days. Assuming that all enterprises are seven-day work system per week, the specific expression is as follows:

[0019]

[0020] In the formula, q r represents the proportion of the load of the non-continuous production type enterprise participating in the week peak-shaving under the level r to the total load gap of the level r, and 0 < q < 1; N con and N dis respectively represent the number of households of the continuous production enterprises and the non-continuous production enterprises among all the enterprises to be involved in the ordered electricity consumption. and respectively represent the cost function of the continuous production enterprise i and the non-continuous production enterprise j for reducing unit load (kW); and respectively represent the load value (kW) of the continuous production enterprise i and the non-continuous production enterprise j participating in the ordered electricity consumption under the level r in the future one week; and respectively represent the electricity value coefficient and the unit electricity production value of the continuous production enterprise i. is a 0-1 variable, which represents whether the continuous production enterprise i participates in the day peak-shaving under the level r, and is equal to 1 when participating; s r % represents the reduction ratio of the continuous production enterprise i under the level r based on the predicted future one-week load, and 0 < s r % < 1; represents the predicted load value (kW) of the continuous production enterprise i in the future one week. and Let represent the electricity value coefficient and unit electricity output value of discontinuous production enterprise j, respectively. Let d represent the number of days the discontinuous enterprise will be shut down in the coming week, d = [1, 7], d ∈ Z; d I represents the set of all possible scenarios where a discontinuous firm is shut down for d days in the coming week. d ={I1,I2,I3,I4,I5,I6,I7};stop d This indicates a possible scenario where a non-continuous business suspends operations for d days in the coming week. It is a 0-1 variable, indicating whether discontinuous production firm j will participate in weekly peak avoidance in the coming week under level r. A value of 1 indicates participation. It is a 0-1 variable, indicating whether non-continuous production enterprise j will stop production for d days under level r in the next week, with 1 indicating yes; It is a 0-1 variable, representing the stop rate of discontinuous production firm j in the coming week under level r. d Whether work has stopped; a value of 1 indicates yes. This represents the predicted load (kW) of discontinuous production enterprise j on day d of the next week.

[0021] The constraints of the model include:

[0022]

[0023]

[0024] In the formula, This indicates the load gap (kW) that the orderly power consumption and peak shaving scheme formulated under level r needs to meet. dD, dC, dB, and dA represent the number of users of discontinuous production enterprises in categories D, C, B, and A, respectively. dD, dC, dB, and dA represent the number of days in the coming week that discontinuous production enterprises in categories D, C, B, and A participate in peak-avoidance activities. These represent the number of users in categories D, C, B, and A for continuous production enterprises, respectively.

[0025] 2) Based on the optimization model results of the orderly power consumption peak-avoidance scheme on a weekly time scale, the list of continuous production enterprises participating in daily peak-avoidance can be obtained. Then, based on the clustering results of the load curves of each group of users, an optimization model of the orderly power consumption peak-avoidance scheme for continuous production enterprises on an hourly time scale is established, thereby providing differentiated solutions for enterprises with different power consumption characteristics. The objective function is to minimize the power outage losses of continuous production enterprises, and the constraint condition is a power balance constraint; the expression of the objective function is as follows:

[0026]

[0027] where day represents the day of the week in the future; p represents the time period of each day in the future week, p=1 represents the morning peak period (7:00-11:00), p=2 represents the waist load period (11:00-17:00), and p=3 represents the evening peak period (17:00-23:00); is a 0-1 variable, obtained from the result of the optimization model of the orderly power utilization peak-shaving scheme under the week time scale; is a 0-1 variable, representing whether the continuous production enterprise i participates in orderly power utilization under the level r on the day day of the future week and in the p time period, and equal to 1 when participating; represents the limited production proportion of the continuous production enterprise i under the level r based on the predicted daily load of the future week; represents the predicted load value (kW) of the continuous production enterprise i on the day day of the future week and in the p time period.

[0028] The constraint condition of the model is the power balance constraint:

[0029] Under each level (A-F level), the sum of the limited production load of the continuous production enterprise in each time period of each day in the future week is equal to the limited production load of the future week obtained under the week time scale, and the expression is as follows:

[0030]

[0031] where s r is obtained from the result of the optimization model of the orderly power utilization peak-shaving scheme under the week time scale, and is the set of continuous production enterprises participating in daily peak-shaving.

[0032] Further, the optimal solution of the established mixed integer nonlinear model (i.e., the orderly power utilization peak-shaving scheme optimization model considering enterprise production characteristics and multi-time scale) is searched based on the branch and bound method, and the orderly power utilization peak-shaving scheme under the A-F six warning levels is obtained.

[0033] The beneficial effects of the present application are:

[0034] The application evaluates the electricity value of enterprises from the aspects of benefits, environmental protection, scientific research and the like. Compared with the existing ordered electricity peak-shaving scheme method which only evaluates the electricity value of enterprises from the benefit index, the evaluation range is more comprehensive, so that the subsequent division of electricity grades of enterprises is more scientific and reasonable. According to the comprehensive score of the electricity value of enterprises, the enterprises are sorted, and then based on the electricity attribute of the enterprises, the enterprises are divided into four groups of key restriction type, general restriction type, general guarantee type and priority guarantee type, and the load of the enterprises is controlled in grades; according to the principle of "graded ordered electricity", the electricity of enterprises with high comprehensive electricity value is preferentially met, and the electricity of enterprises with high energy consumption and high pollution is controlled; according to the principle of "guarantee and pressure", the user of the must guarantee type is not included in the ordered electricity peak-shaving scheme, the electricity scale of the key restriction type enterprise is expanded in the same grade ordered electricity peak-shaving scheme compared with the general restriction type enterprise, and the electricity scale of the guarantee type enterprise is reduced in the same grade ordered electricity peak-shaving scheme compared with the general restriction type enterprise; the production reduction proportion increases upwards according to the ordered electricity peak-shaving scheme A-F grade, and decreases in turn according to the electricity grade. The application considers that the production technology, environment and site required by enterprises of the same production type are different, and then the electricity law characteristics of enterprises are different, so in order to fully consider the individualization of enterprises, according to the clustering results of the daily load curve of continuous production enterprises, the load of the enterprises is controlled in time periods, so that the safe and stable operation of the power grid is realized while the power loss of the enterprises is reduced as much as possible, and the feasibility of the scheme is improved. The ordered electricity peak-shaving scheme formed by the application balances the load reduction capacity of seven days in a week, can match the power gap, and can help and guide to obtain the ordered electricity peak-shaving scheme of enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is a frame diagram of ordered electricity user grouping.

[0036] Figure 2 is a whole flow chart of the application.

[0037] Figure 3 is a clustering result diagram of D-class continuous production enterprises based on K-means.

[0038] Figure 4 is a clustering result diagram of C-class continuous production enterprises based on K-means.

[0039] Figure 5 is a future weekly daily production reduction proportion result diagram of continuous production enterprises participating in ordered electricity under A-F grade.

[0040] Figure 6 is a weekly peak-shaving amount result diagram of continuous production enterprises participating in ordered electricity under A-F grade. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical scheme and advantages of the present application clearer, the technical scheme of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application.

[0042] An ordered power peak-shaving scheme optimization method under A-F grades is provided in the embodiment, as shown in the formula (1), the method comprises the following steps 1-3: Figure 2

[0043] Step 1, according to the load gap under different ordered power peak-shaving scheme grades of the target power grid area in the future one week and the basic information of ordered power of industrial and commercial enterprises, determine the index for evaluating the value of enterprise power, according to the index data, calculate the enterprise power value coefficient and unit power output value;

[0044] The basic information of ordered power of industrial and commercial enterprises includes historical annual load data, predicted future one-week load data, enterprise production mode, enterprise power attribute, annual main business income, enterprise land area, enterprise actual tax payment, enterprise added value, energy consumption, annual average number of employees, enterprise annual output value, research and experimental development fund expenditure, and pollution emission right data. The index for evaluating the value of enterprise power includes per mu tax, per mu added value, per mu output value, unit energy consumption added value, full-time labor rate, unit pollution emission right added value, and research and experimental development fund expenditure proportion of main business income.

[0045] In the embodiment, the load gap under different grades is assumed to be: A grade: 40,000 kW; B grade: 80,000 kW; C grade: 170,000 kW; D grade: 240,000 kW; E grade: 300,000 kW; F grade: 500,000 kW.

[0046] In the embodiment, 25 enterprises in five industries of food and beverage and tobacco product specialized retail, daily use and medical rubber product manufacturing, cotton textile and printing and dyeing finishing, metal processing machinery manufacturing, and property management in Huzhou City, Zhejiang Province are selected. The daily load prediction data of the 25 enterprises from July 12, 2021 to July 18, 2021 is selected as the load data.

[0047] The specific calculation formula of the enterprise user power value coefficient is as follows:

[0048]

[0049] In the formula, S i is the power value coefficient of the i-th enterprise, W g is the contribution of the g-th index in the user power value, P ig is the data of the g-th index of the i-th enterprise.

[0050] The unit power output value is calculated by dividing the annual output value by the annual power consumption. ​

[0051] Step 2, grouping according to the electricity attribute of enterprises, and further grouping the enterprises in each group according to the production mode of enterprises into continuous production enterprises and non-continuous production enterprises, and clustering the continuous production enterprises in each group according to the predicted future weekly daily load similarity by using K-means;

[0052] The framework of the ordered electricity user grouping is shown in Figure 1 According to the electricity attribute of enterprises, four groups are divided, each group is a class, specifically, class A is a priority guarantee class, class B is a general guarantee class, class C is a general restriction class, and class D is a key restriction class.

[0053] In this embodiment, the class D contains 3 non-continuous production enterprises and 3 continuous production enterprises, the class C contains 4 non-continuous production enterprises and 7 continuous production enterprises, the class B contains 4 non-continuous production enterprises, and the class A contains 4 non-continuous production enterprises. The clustering result of the continuous production enterprises in the classes D and C is shown in Figures 3-4

[0054] Step 3, according to the grouping and clustering results, and the electricity value coefficient of enterprises and the unit electricity production value, an ordered electricity peak-shaving scheme optimization model considering the production characteristics of enterprises is established for the weekly and hourly time scales, the branch and bound method is used to solve the optimization model, and the ordered electricity peak-shaving scheme under the A-F six levels of early warning is obtained.

[0055] The ordered electricity peak-shaving scheme optimization model under the weekly time scale is established, and according to the future one-week power gap, the ordered electricity peak-shaving scheme of the continuous production enterprises and the non-continuous production enterprises is determined.

[0056] The objective function of the ordered electricity peak-shaving scheme optimization model under the weekly time scale is to minimize the cost caused by power outages to enterprise users, and the constraint conditions include power balance constraint, corresponding constraints of the number of users, the production limiting proportion of continuous production enterprises, and the shutdown days of non-continuous production enterprises between the ordered electricity peak-shaving schemes under different levels (A-F level). Assuming that the enterprises are seven-day working system, the specific expression of the objective function is as follows:

[0057]

[0058] In the formula, q r represents the proportion of the load of the non-continuous production enterprises participating in the weekly peak-shaving under the level r to the total load gap under the level r, and 0 < q < 1; N con and N dis respectively represent the number of households of the continuous production enterprises and the non-continuous production enterprises among all the enterprises participating in the ordered electricity. and ​respectively, represent the cost function of continuous production enterprise i and non-continuous production enterprise j for reducing unit load (kW); and respectively, represent the load value (kW) of continuous production enterprise i and non-continuous production enterprise j participating in orderly power utilization at level r in the future week. and respectively, represent the power utilization value coefficient and unit power production value of continuous production enterprise i. is a 0-1 variable, representing whether continuous production enterprise i participates in daily peak-shaving at level r, and equals 1 when participating; r s represents the production limiting ratio of continuous production enterprise i at level r based on the predicted load in the future week, and 0 r < s represents the predicted load value (kW) of continuous production enterprise i in the future week. and respectively, represent the power utilization value coefficient and unit power production value of non-continuous production enterprise j. d represents the number of days of non-continuous enterprise stopping in the future week, d = [1, 7], d ∈ Z; d represents the set of all possible cases of which days the non-continuous enterprise stops when stopping for d days in the future week, d = {I1, I2, I3, I4, I5, I6, I7}; stop d represents a possible case of non-continuous enterprise stopping for d days in the future week; is a 0-1 variable, representing whether non-continuous production enterprise j participates in weekly peak-shaving at level r in the future week, and equals 1 when participating; represents whether non-continuous production enterprise j stops for d days at level r in the future week, and equals 1 when yes; represents whether non-continuous production enterprise j stops at level r in the future week; d represents the predicted load value (kW) of non-continuous production enterprise j on the dth day in the future week.

[0059] The constraint conditions of the model include:

[0060]

[0061]

[0062] In the formula, represents the load gap (kW) that needs to be met by the orderly power utilization peak-shaving scheme formulated at level r. ​respectively represent the number of users of non-continuous production enterprises in D, C, B, and A classes. respectively represent the number of users of non-continuous production enterprises in D, C, B, and A classes.

[0063] According to the results of the optimization model of the orderly electricity peak-shaving scheme under the weekly time scale, the list of continuous production enterprises participating in the daily peak-shaving is obtained, and according to the clustering results of the load curve of each group of users, an optimization model of the orderly electricity peak-shaving scheme of continuous production enterprises under the hourly time scale is established to optimize the load limiting production proportion of continuous production enterprise users in the early peak, waist load, and late peak period of each day in the future week.

[0064] In the optimization model of the orderly electricity peak-shaving scheme under the hourly time scale, the objective function is to minimize the power outage cost of the continuous production enterprise, and the constraint condition is the power balance constraint; the expression of the objective function of the model is as follows:

[0065]

[0066] In the formula, day represents the day in the future week; p represents the time period in each day in the future week, p=1 represents the early peak period (7:00-11:00), p=2 represents the waist load period (11:00-17:00), and p=3 represents the late peak period (17:00-23:00); is a 0-1 variable, which is obtained from the results of the optimization model of the orderly electricity peak-shaving scheme under the weekly time scale; is a 0-1 variable, which represents whether the continuous production enterprise i participates in the orderly electricity in the future week of day and p under the level r, and is equal to 1 when participating; represents the limiting production proportion of the continuous production enterprise i based on the predicted daily load in the future week under the level r; represents the predicted load value (kW) of the continuous production enterprise i in the future week of day and p.

[0067] The constraint condition of the model is the power balance constraint:

[0068] Under each level (A-F level), the sum of the limiting production load of the continuous production enterprise in each period of each day in the future week is equal to the limiting production load of the future week obtained under the weekly time scale, and the expression is as follows:

[0069]

[0070] In the formula, s r% is obtained by the result of the optimization model of the peak-shaving scheme of the orderly power utilization under the weekly time scale, and is the set of continuous production enterprises participating in the daily peak-shaving.

[0071] In this embodiment, the load clustering result of the D-class and C-class continuous production enterprises based on Kmeans is as shown in Figures 3-4 The daily load prediction data of the continuous production enterprises on July 12 (Monday) is clustered. For the continuous production enterprises with high energy consumption (such as enterprise 4 and 8 shown in Figure 4 ), it is set to participate in the early peak, waist load and late peak period, and the specific production limiting proportion in each period is determined according to the objective function. For the continuous production enterprises with double-peak load curve (such as enterprise 3 shown in Figure 3 ), it is set to participate in the early peak and late peak period, and the specific production limiting proportion in each period is determined according to the objective function. For the continuous production enterprises with low load rate (such as enterprise 1 and 2 shown in Figure 3 and enterprise 5, 6, 7, 9 and 10 shown in Figure 4 ), the specific participating period and the corresponding production limiting proportion in each period are determined according to the objective function.

[0072] Step 3.3, based on the branch and bound method, the optimal solution of the established mixed integer nonlinear model is searched, and the orderly power utilization peak-shaving scheme under A-F grade early warning is obtained.

[0073] Among them, the orderly power utilization peak-shaving scheme under different grades in this embodiment is:

[0074] 1) Under A grade, D-class non-continuous production enterprises stop one and start six, and D-class continuous production enterprises have a weekly load limiting proportion of 3.58%.

[0075] 2) Under B grade, D-class non-continuous production enterprises stop two and start five, and D-class continuous production enterprises have a weekly load limiting proportion of 6.45%.

[0076] 3) Under C grade, D-class non-continuous production enterprises stop three and start four, C-class non-continuous production enterprises stop one and start six, and D-class and C-class continuous production enterprises have a weekly load limiting proportion of 10.5%.

[0077] 4) Under D grade, D-class non-continuous production enterprises stop four and start three, C-class non-continuous production enterprises stop two and start five, and D-class and C-class continuous production enterprises have a weekly load limiting proportion of 11.5%.

[0078] 5) Under E level, D type non-continuous production enterprises stop four open three, C type non-continuous production enterprises stop three open four, B type non-continuous production enterprises stop one open six, D type and C type continuous production enterprises week load limit production ratio 15.9%.

[0079] 6) Under F level, D type non-continuous production enterprises stop five open two, C type non-continuous production enterprises stop four open three, B type non-continuous production enterprises stop two open five, D type and C type continuous production enterprises week load limit production ratio 44.1%.

[0080] The power-off cost corresponding to the week peak avoidance measures of the method and the traditional method is shown in Table 1. Compared with the traditional method, the method can greatly reduce the power-off cost. The non-continuous production enterprises stop working which day under different levels in the future week is shown in Tables 2-7, and the limit production ratio and peak avoidance load of continuous production enterprises in the early peak, waist load and late peak period every day are shown in Tables 8-11. Figure 5 And Figure 6 .

[0081] Table 1 Comparison of power-off cost corresponding to week peak avoidance measures of the method and the traditional method

[0082]

[0083] Table 2 Week peak avoidance measures of non-continuous production enterprises under A level

[0084]

[0085]

[0086] Table 3 Week peak avoidance measures of non-continuous production enterprises under B level

[0087]

[0088] Table 4 Peak avoidance measures of non-continuous production enterprises under C level

[0089]

[0090] Table 5 Week peak avoidance measures of non-continuous production enterprises under D level

[0091]

[0092] Table 6 Week peak avoidance measures of non-continuous production enterprises under E level

[0093]

[0094] Table 7 Week peak avoidance measures of non-continuous production enterprises under F level

[0095]

[0096]

[0097] The above-described embodiments are only used to illustrate the technical solutions of the present application but not to limit the present application. Although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still make modifications to the technical solutions of the foregoing embodiments, or make equivalent replacements to some of the technical features without departing from the spirit and scope of the present application, and any modifications or equivalent replacements should be included in the protection scope of the claims of the present application.

Claims

1. An ordered electricity consumption peak avoidance scheme optimization method considering enterprise production characteristics and multi-time scales, characterized in that, The method comprises the following steps: Step 1, determining indexes for evaluating the electricity value of enterprises according to the load gap of different orderly electricity consumption peak-shaving scheme levels in a target power grid area in the future week and the basic information of orderly electricity consumption of industrial and commercial enterprises, and calculating the electricity value coefficient and unit electricity production value of enterprises according to the index data; Step 2, grouping the enterprises according to the electricity properties of the enterprises, and then further grouping the enterprises in each group into continuous production enterprises and non-continuous production enterprises according to the production modes of the enterprises, and clustering the continuous production enterprises in each group by using the K-means method according to the predicted daily load similarity in the future week; Step 3, establishing an orderly electricity consumption peak-shaving scheme optimization model considering the production characteristics and multi-time scales of enterprises according to the grouping and clustering results and the electricity value coefficient and unit electricity production value of the enterprises, and solving the model to obtain orderly electricity consumption peak-shaving schemes in several warning levels; In step 3, the orderly electricity consumption peak-shaving scheme optimization model considering the production characteristics and multi-time scales of enterprises comprises two stages, and specifically comprises the following steps: 1) establishing an optimization model of the orderly electricity consumption peak-shaving scheme in the week time scale; the objective function of the optimization model of the orderly electricity consumption peak-shaving scheme in the week time scale is to minimize the cost caused by power failure to enterprise users, and the constraint conditions include power balance constraint, corresponding constraints of the number of users, the production limiting proportion of continuous production enterprises and the shutdown days of non-continuous production enterprises in different levels of orderly electricity consumption peak-shaving schemes; 2) obtaining the list of continuous production enterprises participating in daily peak-shaving according to the solving result of the optimization model of the orderly electricity consumption peak-shaving scheme in the week time scale, and establishing an optimization model of the orderly electricity consumption peak-shaving scheme of continuous production enterprises in the hour time scale according to the clustering result of the load curve of each group of users; the objective function of the optimization model of the orderly electricity consumption peak-shaving scheme of continuous production enterprises in the hour time scale is to minimize the power failure cost of continuous production enterprises, and the constraint condition is the power balance constraint.

2. The orderly electricity consumption peak-shaving scheme optimization method considering the production characteristics and multi-time scales of enterprises according to claim 1, characterized in that: In step 1, the basic information of orderly electricity consumption of industrial and commercial enterprises includes historical annual load data, predicted daily load data in the future week, production mode of enterprises, electricity property of enterprises, annual main business income, enterprise land area, enterprise real tax payment, enterprise added value, energy consumption, annual average number of employees, enterprise annual output value, research and development expenditure, and pollution discharge right data; the indexes for evaluating the electricity value of enterprises include per mu tax, per mu added value, per mu output value, unit energy consumption added value, full-time labor rate, unit pollution discharge right added value, and proportion of research and development expenditure to main business income; The calculation formula of the electricity value coefficient of enterprises is as follows: Wherein, S i is the electricity value coefficient of the i-th enterprise, W g is the contribution of the g-th index in the electricity value of the user, P ig is the data of the g-th index of the i-th enterprise; The unit electricity production value is calculated by dividing the annual output value of the enterprise by the annual electricity consumption.

3. The orderly electricity consumption peak-shaving scheme optimization method considering the production characteristics and multi-time scales of enterprises according to claim 1, characterized in that: In step 2, the enterprises are divided into four groups according to the enterprise electricity properties, and each group is a class, specifically, class A is a priority guarantee class, class B is a general guarantee class, class C is a general restriction class, and class D is a key restriction class; when calling the user of the orderly electricity peak-shaving scheme, the users are called in the order of class D, class C, class B and class A.

4. The orderly electricity peak-shaving scheme optimization method considering enterprise production characteristics and multiple time scales according to claim 1, characterized in that: The optimization model of the orderly electricity peak-shaving scheme under the weekly time scale is specifically: Assuming that all enterprises have a seven-day work system every week, the specific expression of the objective function is as follows: In the formula, z represents the number of warning levels r, i.e., r = 1, 2, ..., z; q r This represents the proportion of load that a discontinuous production enterprise participates in during peak shaving at level r, relative to the total load gap at level r, where 0 < q < 1; N con and N dis These represent the number of continuous production enterprises and non-continuous production enterprises among all enterprises awaiting participation in the orderly electricity consumption program; and Let i represent the cost function for reducing the unit load kW of continuous production enterprise i and discontinuous production enterprise j, respectively. and These represent the load values ​​(in kW) of continuous production enterprise i and discontinuous production enterprise j participating in orderly electricity consumption under level r in the coming week; and Let represent the electricity value coefficient and the output value per unit of electricity for continuous production enterprise i, respectively; It is a 0-1 variable, representing whether continuous production firm i participates in daily peak shaving at level r; a value of 1 indicates participation. r % represents the percentage of production cuts by continuous production firm i at level r based on the projected load for the next week, and 0 < s r % < 1; P i cw This represents the predicted load value (in kW) of continuous production enterprise i in the coming week; and Let represent the electricity value coefficient and unit electricity output value of discontinuous production enterprise j, respectively; d represents the number of days the discontinuous enterprise will be shut down in the next week, d = [1, 7], d ∈ Z; I d I represents the set of all possible scenarios where a discontinuous firm is shut down for d days in the coming week. d ={I1,I2,I3,I4,I5,I6,I7};stop d This indicates a possible scenario where a non-continuous business suspends operations for d days in the coming week. It is a 0-1 variable, indicating whether discontinuous production firm j will participate in weekly peak avoidance in the coming week under level r. A value of 1 indicates participation. It is a 0-1 variable, indicating whether non-continuous production enterprise j will stop production for d days under level r in the next week, with 1 indicating yes; It is a 0-1 variable, representing the stop rate of discontinuous production firm j in the coming week under level r. d Whether work has stopped; a value of 1 indicates yes. This represents the predicted load value (in kW) of discontinuous production enterprise j on day d of the next week; The constraint conditions of the model include: s F %>s E %>s D %>s C %>s B %>s A % (13) In the formula, represents the load gap kW that needs to be met by the ordered peak-shaving scheme formulated at the level r; respectively represent the number of users of non-continuous production enterprises in D, C, B, and A classes; dD, dC, dB, and dA respectively represent the number of days in the future week that the non-continuous production enterprises in D, C, B, and A classes participate in weekly peak-shaving; respectively represent the number of users of continuous production enterprises in D, C, B, and A classes; The optimization model of the orderly electricity peak-shaving scheme of the continuous production enterprise with an hourly time scale is specifically: The expression of the objective function is as follows: where day represents the day of the week in the future; p represents the time period of the day in the future week: p = 1 represents the morning peak period, 7:00-11:00; p = 2 represents the waist load period, 11:00-17:00; p = 3 represents the evening peak period, 17:00-23:00; is a 0-1 variable, indicating whether the continuous production enterprise i participates in orderly power utilization at the level r on the day day in the future week and the p time period, and equal to 1 indicates participation; represents the limited production proportion of the continuous production enterprise i at the level r on the basis of the daily load of the predicted future week, represents the predicted load value kW of the continuous production enterprise i on the day day in the future week and the p time period. The constraint conditions of the model include: In the formula, is a collection of enterprises that participate in the continuous production type of day peak avoidance.

5. The method of claim 4, wherein the method further comprises: The optimal solution of the orderly electricity peak-shaving scheme optimization model considering enterprise production characteristics and multiple time scales is searched by using the branch and bound method, and the orderly electricity peak-shaving schemes under different levels of early warning are obtained.

Images