An industrial enterprise power demand monitoring method and system

Abstract

The application discloses an industrial enterprise power demand monitoring method and system, which comprises the following steps: pre-grouping the enterprise adjustable load according to the power data and priority; continuously calculating the demand in the demand time period; calculating the power limit value of the power demand measurement point and the average active power limit value of the future time period; grouping the demand time period, calculating the average active power limit value of multiple future time period groups, and step by step adjusting the adjustable load ladder; calculating the adjustable capacity of each group of load and the total adjustable capacity, and issuing the adjustment instruction in the group time period; storing all data in real time; dividing the adjustable load in the enterprise into k clusters, and optimizing the division of multiple time period groups; after the monitoring cycle of the current month demand is ended, the maximum value of the current month demand is calculated, which is used as the basis for signing the demand limit value with the power grid by the enterprise in the next month. The application provides an economic and reliable power demand management and control solution for the enterprise, and has important application value for promoting energy saving, emission reduction, cost reduction and benefit increase of the enterprise.

Description

Technical Field

[0001] This invention belongs to the field of power systems for industrial enterprises, and specifically relates to a method and system for monitoring the power demand of industrial enterprises. Background Technology

[0002] Electricity demand refers to the average active power over a specified time interval, while monthly maximum demand is the maximum electricity demand within that month, measured in kilowatts (kW). Currently, my country uses a two-part pricing system for electricity billing for large industrial users. This system includes a fixed monthly charge based on the transformer capacity or maximum demand of the industrial enterprise. The basic charge is calculated based on a contract between the power supply and user departments, specifying a fixed limit that is not affected by actual electricity consumption. In some provinces and cities, monthly real-time demand-based billing has been implemented. The electricity charge is calculated based on the actual electricity consumption of the user. Using "maximum demand" as the basic charge helps enterprises rationally manage their load operations, minimizing the simultaneous operation of multiple peak load demand intervals. By controlling and regulating the adjustable load, maximum demand can be reduced, thereby lowering production costs.

[0003] In the prior art, Chinese invention patent with publication number CN106372451A (hereinafter referred to as prior art 1) discloses a method for calculating electricity demand. This method first obtains the load information of each demand metering point in the current month, then obtains the load information of the demand metering point with the maximum demand value in the reference month as a reference sample, then finds the demand metering point with the smallest transfer change rate that can be connected to implement load transfer, then updates the historical load value of the reference month, and then calculates the daily maximum load of each sequence metering point in the reference month after load transfer; then calculates the monthly maximum load value of each demand metering point in the reference month, and then calculates the actual maximum demand that can be declared in the next cycle based on the calculation results. Chinese invention patent CN111342459A (hereinafter referred to as Prior Art 2) discloses a power demand decision analysis system and method. The system includes a calculation experiment unit, an evaluation analysis unit, and a control strategy execution unit connected in sequence, as well as a multi-source data calculation unit and a management and control constraint unit. The two ends of the multi-source data calculation unit are connected to the control strategy execution unit and the calculation experiment unit, respectively. The two ends of the management and control constraint unit are connected to the evaluation analysis unit and the calculation experiment unit, respectively. The external feedback loop, consisting of the calculation experiment unit, the evaluation analysis unit, the control strategy execution unit, and the multi-source data calculation unit, realizes the calculation of demand and the early warning of exceeding limits. The internal feedback loop, consisting of the calculation experiment unit, the evaluation analysis unit, and the management and control constraint unit, continuously optimizes the calculation of the demand exceeding limits at each threshold by combining the demand change trend and historical data at each threshold.

[0004] However, the existing technologies mentioned above have the following shortcomings: 1) Existing technology 1 has two or more grid connection points, which means there are two or more demand metering points. The load can be transferred between the grid connection points. It can be seen that when the load is transferred to different demand metering points, two or more demand metering points are required. However, most enterprises only have one demand metering point, so this method cannot be used. In addition, the load will start and stop during the load transfer, which is not suitable for frequent operations; 2) Although the demand calculation adopts the slip method, the demand period and slip interval are not adjustable. The slip calculation is performed for the entire time period, without considering the unchanging intermediate time period; 3) Existing technology 2 only monitors and controls the installation device at the demand metering point. It cannot establish an enterprise-level adjustable load control model and does not establish a group optimization and rapid control of adjustable loads. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a method and system for monitoring the power demand of industrial enterprises, which can overcome the defects of existing technologies such as the inability to transfer load and the repetitive calculation of demand.

[0006] Technical solution: The present invention provides a method for monitoring the electricity demand of industrial enterprises, comprising:

[0007] The power data of the load is obtained through the enterprise's power centralized control system, and the enterprise's adjustable loads are pre-grouped according to the adjustment priority based on the power data;

[0008] The demand time period is calculated using the rolling slip calculation method. The demand within the period is calculated on a rolling basis using a slip time scale, where the slip time scale and the demand period are... Adjust according to demand;

[0009] Demand time period The calculation is divided into two time periods: the period that has already occurred and the period that will occur in the future. The rolling slip calculation method is used to calculate the electricity consumption during the period that has already occurred. and demand time period Battery During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. Average active power over the future time period Less than the average active power limit for the future time period Proceed to the next step if the conditions are met;

[0010] Divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the first... The group has experienced a period of time in terms of electricity consumption. Combined with the electricity demand metering point's electricity limit and the The group has experienced a period of time in terms of electricity consumption. Calculate the first Average active power limit for future time period The stepwise approximation method is adopted, based on the first... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in stages, in the first stage... Average active power of the group in the future time period Less than In this case, further determine the real-time active power at the demand metering point. Is it less than ;

[0011] If the demand metering point has real-time active power Exceed Based on the pre-grouping of adjustable loads, the grouped active power of each adjustable load and the total active power of all adjustable loads are obtained; combined with the grouped active power of each adjustable load, the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. According to the first Adjustable capacity of load group Total adjustable capacity of adjustable load Adjustment commands are issued to the adjustable load in groups and time periods for adjustment;

[0012] The system stores the adjustment status of all the above data and adjustment commands in real time.

[0013] The K-means clustering algorithm, based on artificial intelligence, is used to divide the adjustable load within the enterprise into k clusters according to the stored data, thereby grouping the adjustable load. After grouping, the real-time power of the demand metering points after adjustment is determined. If the active power is less than the average active power limit for multiple future time periods, then increase the regulating load and readjust.

[0014] After the demand monitoring period ends this month, the maximum demand for this month will be calculated based on historical data, which will serve as the basis for enterprises to sign demand limit agreements with the power grid next month.

[0015] Furthermore, the rolling slip calculation method is used to continuously calculate the demand within a demand period, performing rolling calculations on a slip time scale, wherein the slip time scale and the demand period are... Adjustments will be made according to demand, including:

[0016] Assuming the demand period is 15 minutes, starting from the first minute of the month, the active power is integrated from 0 to 15 minutes to obtain the electricity consumption for the first 15-minute period. The electricity consumption for this 15-minute period is divided by the time period to obtain the average active power for this period. When the time reaches the 16th minute, the electricity consumption for the first 1 minute is automatically subtracted, and the electricity consumption for the first 15 minutes is automatically added to the total electricity consumption for the new 15-minute period. The average active power for the new time period is also calculated, as shown in the following formula.

[0017] ;

[0018] In the formula, From Battery level for the next 15 minutes from this time; From Battery level for 1 minute from the current time; for Battery level one minute prior to the event; From The average active power over the next 15 minutes from that time; Real-time active power at the demand metering point; From The average active power over the next 15 minutes from that time; The start time for demand calculation; This is the end time for demand calculation.

[0019] Furthermore, the required time period The calculation is divided into two time periods: the period that has already occurred and the period that will occur in the future. The rolling slip calculation method is used to calculate the electricity consumption during the period that has already occurred. and demand time period Battery ,include:

[0020] Demand time period Divided into two time periods: already occurred and future. , Using the equal area method, calculate the electricity values ​​for the past and future time periods:

[0021] ;

[0022] In the formula, This represents the amount of electricity consumed during the time period in question. This is the current calculation time; This refers to the start time of a past time period. For future energy consumption; This refers to the end time of a future time period. This represents the real-time active power value collected at the demand metering point. Demand period Battery level.

[0023] Furthermore, the electricity limit combined with the electricity demand metering point Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. ,include:

[0024] If the electricity demand limit is the maximum average active power within the demand time period T, then the expression for the electricity demand limit at the metering point is as follows:

[0025]

[0026] In the formula, Demand limits for electricity demand metering points; The electricity consumption limit for electricity demand metering points; For the time period of demand;

[0027] Combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. The expression is as follows:

[0028] .

[0029] Furthermore, the past and future time periods are divided into multiple groups, each group including both past and future time periods, and the electricity consumption of the i-th group of past time periods is calculated. ,include:

[0030] Demand time period Divided into multiple groups, each representing a past event and a future event, respectively. , ... ,in, For the first Group; using the method of equal area of ​​electric charge, calculate the first group; The electricity consumption values ​​for the two time periods are shown in the following formula:

[0031] ;

[0032] In the formula, For the first Group demand for electricity during a specific time period; For the first The group has consumed electricity over the specified period of time; This is the current calculation time; For the first The start time of the past time period; For the first Group the amount of electricity to be generated in the future time period. For the first The end time of the future time period.

[0033] Furthermore, the first The expression for the average active power limit for the future time period is as follows:

[0034] ;

[0035] In the formula, For the first The average active power limit for the group over a future time period; The electricity consumption limit for electricity demand metering points; For the first The group has consumed electricity over the specified period of time; For the first Group of future time periods.

[0036] Furthermore, the calculation methods for the grouped active power of each adjustable load and the total active power of all adjustable loads are as follows:

[0037] ;

[0038] In the formula, For the first The sum of the active power of the load groups; For the first The first in the group load Active power of each load; This represents the total active power of the load.

[0039] Furthermore, the active power of each adjustable load group is used to calculate the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. The expression is as follows:

[0040] ;

[0041] In the formula, For the first Adjustable capacity of the load group; For the first The first in the group load The coefficient of active power of each load; The total adjustable capacity of the adjustable load; For the first Adjustable coefficient of active power of group load; This is the total adjustability coefficient for adjustable loads.

[0042] Furthermore, the K-means clustering algorithm, based on stored data, is used to divide the adjustable load within the enterprise into k clusters, thereby grouping the adjustable load, including:

[0043] The stored data is preprocessed, including the removal of outlier data and the human intervention of load importance levels;

[0044] By selecting an appropriate value of k using the Elbow Method, each adjustable load is assigned to the nearest group center, forming k groups.

[0045] Run the k-means algorithm to recalculate the center of each group, that is, the mean of all points in each group;

[0046] Determine whether the clustering has converged or reached the desired number of iterations. If not, repeat the above steps until the group centers no longer change or the preset number of iterations is reached, and then output the clustering results.

[0047] Analyze the calculation results and examine the characteristics of different groups;

[0048] The K-means clustering algorithm for artificial intelligence runs in real time and feeds back the grouping results to the demand analysis and control algorithm in real time.

[0049] Based on the same inventive concept, the present invention provides an industrial enterprise power demand monitoring system, comprising:

[0050] The power centralized control data acquisition and processing module is used to acquire the power data of the load through the enterprise's power centralized control system, and pre-group the enterprise's adjustable loads according to the adjustment priority based on the power data;

[0051] The electricity demand calculation and analysis module is used to calculate the demand time period using the rolling slip calculation method. The demand within the period is calculated on a rolling basis using a slip time scale, where the slip time scale and the demand period are... Adjust according to demand;

[0052] The electricity demand calculation and analysis module is used to calculate the demand time period. The calculation is divided into two time periods: the period that has already occurred and the period that will occur in the future. The rolling slip calculation method is used to calculate the electricity consumption during the period that has already occurred. and demand time period Battery During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. Average active power over the future time period Less than the average active power limit for the future time period Proceed to the next step if the conditions are met;

[0053] The electricity demand calculation and analysis module is used to divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the demand for electricity. The group has experienced a period of time in terms of electricity consumption. Combined with the electricity demand metering point's electricity limit and the The group has experienced a period of time in terms of electricity consumption. Calculate the first Average active power limit for future time period The stepwise approximation method is adopted, based on the first... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in a step-by-step manner, with the average active power of the i-th group in the future time period. Less than In this case, further determine the real-time active power at the demand metering point. Is it less than ;

[0054] The power demand control command issuing module is used to control the real-time active power at the demand metering point. Exceed Based on the pre-grouping of adjustable loads, the grouped active power of each adjustable load and the total active power of all adjustable loads are obtained; combined with the grouped active power of each adjustable load, the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. According to the first Adjustable capacity of load group Total adjustable capacity of adjustable load Adjustment commands are issued to the adjustable load in groups and time periods for adjustment;

[0055] The power demand control command issuing module is used to store all the above data and the adjustment status of the adjustment commands in real time.

[0056] The AI-based K-means clustering algorithm module is used to divide the adjustable load within an enterprise into k clusters based on stored data, thereby grouping the adjustable load. After grouping, it determines the real-time power of the demand metering points after adjustment. If the active power is less than the average active power limit for multiple future time periods, then increase the regulating load and readjust.

[0057] The data statistics module is used to calculate the maximum demand for the month based on historical data after the end of the monthly demand monitoring cycle, which serves as the basis for enterprises to sign demand limit agreements with the power grid in the following month.

[0058] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are as follows: (1) The data acquisition of the present invention is based on the enterprise power centralized control system, and no longer uses load transfer, no longer separately collects the telemetry and teleindication information of demand metering points and enterprise power systems, and adopts real-time library interface technology, which improves the timeliness of real-time data and overcomes the defect that the load cannot be transferred in the prior art; (2) The demand calculation adopts the rolling slip method, which considers the unchanging intermediate time period. Compared with the arithmetic method, it ensures that the demand value of each demand time period is monitored and calculated. Compared with the conventional slip method, it achieves the effect of adjustable demand calculation time period and slip period, saves system calculation time, and improves system calculation efficiency; (3) The equal power area is adopted, which utilizes the power of the time period and the demand. The maximum power limit value is achieved by accurately calculating the maximum power limit value and active power limit value for future time periods, thus improving the accuracy of control limits; (4) The step-by-step approximation method is adopted to change the calculation time of the time period that has occurred, set multiple time periods, calculate the maximum power limit value and active power limit value for multiple future time periods, obtain the demand control execution strategy, achieve the effect of reducing the maximum demand value, and improve the timeliness of demand control; (5) Artificial intelligence algorithm is adopted to train historical data samples and dynamically group the enterprise's adjustable load, thereby achieving the effect of optimizing load control and improving the flexibility of load group control; (6) The hardware and software in the system of this invention adopt redundant configuration, which improves the system's operational safety, stability and reliability. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating a method for monitoring the electricity demand of an industrial enterprise, as disclosed in an embodiment of the present invention.

[0060] Figure 2 This is a schematic flowchart of a rolling slip calculation method disclosed in an embodiment of the present invention;

[0061] Figure 3 This is a schematic diagram of a method for calculating the area of ​​equal electrical charge disclosed in an embodiment of the present invention;

[0062] Figure 4 This is a flowchart of an adjustable load grouping algorithm based on an artificial intelligence K-means algorithm disclosed in an embodiment of the present invention;

[0063] Figure 5 This is a schematic diagram of the structure of an industrial enterprise power demand monitoring system disclosed in an embodiment of the present invention. Detailed Implementation

[0064] The technical solution of the present invention will now be described in detail with reference to specific embodiments and accompanying drawings.

[0065] Example 1

[0066] like Figure 1 As shown, the present invention provides a method for monitoring the electricity demand of industrial enterprises, which specifically includes the following steps:

[0067] S1. Obtain the power data of the load through the enterprise's power centralized control system, and pre-group the enterprise's adjustable loads according to the adjustment priority based on the power data.

[0068] In this embodiment, power data collected by the enterprise's power centralized control system is acquired in real time through a real-time database interface. The system monitors the active power, voltage, current, and switch status of the loads, pre-groups adjustable loads according to priority, and monitors the regulation margin and regulation rate of each group in real time to evaluate the overall regulation margin and regulation rate.

[0069] S2. Calculate the demand time period using the rolling slip calculation method. The demand within the period is calculated on a rolling basis using a slip time scale, where the slip time scale and the demand period are... Adjustments will be made according to requirements. Details are as follows:

[0070] like Figure 2 As shown, this invention uses a "rolling slip" calculation method. Taking a 15-minute demand period as an example, starting from the first minute of the month, the active power is integrated from 0 to 15 minutes to obtain the electricity consumption for the first 15-minute period. The average active power for this period is obtained by dividing the electricity consumption for this 15-minute period by the time period, which is the demand. When the time reaches the 16th minute, the electricity consumption for the first 1 minute is automatically subtracted, and the electricity consumption for the first 15 minutes is automatically added to the total electricity consumption for the new 15-minute period. The average active power for the new time period is also calculated, as shown in the following formula.

[0071] (1)

[0072] In the formula, From Battery level for the next 15 minutes from this time; From Battery level for 1 minute from the current time; for Battery level one minute prior to the event; From The average active power over the next 15 minutes from that time; Real-time active power at the demand metering point; From The average active power over the next 15 minutes from that time; The start time for demand calculation; This is the end time for demand calculation.

[0073] It is worth noting that in the rolling slip calculation method, the power consumption in the middle time period remains unchanged, the slip period at the beginning of the demand time period is automatically rolled out, and the power consumption in the latest slip period period is automatically rolled in, so as to obtain the average power consumption and active power in the new demand time period. The demand time period and the rolling slip period period can be set.

[0074] S3, Demand Time Period The calculation is divided into two time periods: the period that has already occurred and the period that will occur in the future. The rolling slip calculation method is used to calculate the electricity consumption during the period that has already occurred. and demand time period Battery During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. Average active power over the future time period Less than the average active power limit for the future time period Proceed to the next step if the conditions are met.

[0075] The specific implementation process of step S3 is as follows:

[0076] S3.1, Demand Time Period Dividing the data into two time periods, already occurred and in the future, the electricity consumption for the already occurred time period is calculated using the rolling slip calculation method described above. and demand time period Battery level. Details are as follows:

[0077] Demand time period Divided into two time periods: already occurred and future. , According to the method for calculating electricity consumption using formula (1), the electricity consumption for the past time period and the future time period is calculated:

[0078] (2)

[0079] In the formula, This represents the amount of electricity consumed during the time period in question. This is the current calculation time; This refers to the start time of a past time period. For future energy consumption; This refers to the end time of a future time period. This represents the real-time active power value collected at the demand metering point. Demand period Battery level.

[0080] S3.2, During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. The details are as follows:

[0081] Electricity demand limit is the demand time period Given the maximum average value of internal active power, the expression for the electricity limit at the electricity demand metering point is as follows:

[0082] (3)

[0083] In the formula, Demand limits for electricity demand metering points; The electricity consumption limit for electricity demand metering points; This refers to the demand period. The demand limit at each electricity demand metering point is the demand limit stipulated in the electricity demand fee agreement between the enterprise and the power grid. To ensure that the demand value does not exceed the limit, any demand period... Battery The following conditions must be met:

[0084] (4)

[0085] Electricity consumption during any given demand period satisfy In this case, combined with the electricity demand metering point's electricity limit The average active power limit for future time periods is calculated by combining the electricity consumption during the past time period. Its expression is as follows:

[0086] (5)

[0087] S3.3, Average active power over a future time period Less than the average active power limit for the future time period Proceed to the next step if the above conditions are met. Details are as follows:

[0088] The average power over the future time period is obtained from equation (1):

[0089] (6)

[0090] in, This is the real-time value of active power; The current moment; This refers to the end time of a future period. It is necessary to ensure the supply of goods during the desired time period. Domestic demand should not exceed the limit, and the average active power over the future period. The following conditions must be met:

[0091] (7)

[0092] S4. Divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the... The group has experienced a period of time in terms of electricity consumption. Combined with the electricity demand metering point's electricity limit and the The group has experienced a period of time in terms of electricity consumption. Calculate the first Average active power limit for future time period The stepwise approximation method is adopted, based on the first... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in stages, in the first stage... Average active power of the group in the future time period Less than In this case, further determine the real-time active power at the demand metering point. Is it less than .

[0093] The specific implementation process of step S4 is as follows:

[0094] S4.1 Divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the first... The group has experienced a period of time in terms of electricity consumption. The details are as follows:

[0095] Demand time period Divided into multiple groups, each representing a past event and a future event, respectively. , ... ,in, For the first Group; combining the method of equal area of ​​electric charge, the first group is calculated using formula (2). The electricity consumption values ​​for the two time periods are shown in the following formula:

[0096] (8)

[0097] In the formula, For the first Group demand for electricity during a specific time period; For the first The group has consumed electricity over the specified period of time; This is the current calculation time; For the first The start time of the past time period; For the first Group the amount of electricity to be generated in the future time period. For the first The end time of the future time period.

[0098] S4.2, Electricity limits combined with electricity demand metering points and the electricity consumption during the time period of the i-th group The first number is calculated using formula (5). Average active power limit for future time period Its expression is as follows:

[0099] (9)

[0100] In the formula, For the first The average active power limit for the group over a future time period; The electricity consumption limit for electricity demand metering points; For the first The group has consumed electricity over the specified period of time; For the first Group of future time periods.

[0101] S4.3. Using the stepwise approximation method, according to the... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in stages, in the first stage... Average active power of the group in the future time period Less than In this case, further determine the real-time active power at the demand metering point. Is it less than The details are as follows:

[0102] Using a stepwise approximation method, the changing trends of active power limits over multiple future time periods are obtained. To ensure that the demand value during the demand period T does not exceed the limit, the following formula (6) is used to determine the appropriate limit. Average active power of the group in the future time period All of the following must be satisfied:

[0103] (10)

[0104] exist In this case, As a criterion for real-time power adjustment, the above average active power limits for multiple future time periods and the real-time active power at the demand metering point are used. Compare and determine the real-time active power at the demand metering point. Is it less than This checks whether there is a risk of exceeding limits in multiple future time periods; specifically as follows: if the real-time active power of the demand metering point... If this condition is met, then there is no risk of exceeding the limit in multiple future time periods; otherwise, then... If a limit is exceeded, proceed to the next step.

[0105] S5. If the active power at the metering point is real time Exceeding the Average active power limit for future time period Based on the pre-grouping of adjustable loads, the real-time active power of each group of adjustable loads and the total active power of all adjustable loads are obtained. Combined with the grouped active power of each group of adjustable loads, the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. According to the first Adjustable capacity of load group Total adjustable capacity of adjustable load Adjustment commands are issued to the adjustable load in groups and time periods for adjustment.

[0106] Based on the adjustable load groups and their priorities, adjust each adjustable load sequentially until the real-time active power at the metering point is reached. Average active power limits for multiple future time periods No limit exceedances were detected after comparison. The adjustment methods for each adjustable load are as follows:

[0107] Enterprise adjustable load grouping management divides load resources such as production equipment and auxiliary systems into high / medium / low importance control groups based on their adjustable characteristics (such as response speed and interruption tolerance). Combined with the aforementioned multi-group segmented power limits (e.g., 30% load reduction for air conditioning systems under normal conditions, 50% power restriction for auxiliary systems in emergencies, and 80% power restriction for production lines under special circumstances), dynamic adjustment commands are issued. Power limit control is prioritized for low importance groups (such as discontinuous equipment), while high importance groups (such as core production lines) are allowed to operate within safe thresholds. This achieves refined control through "group response - limit constraints - time-period matching," ultimately achieving the dual goals of optimizing enterprise energy demand costs and ensuring safe production.

[0108] By pre-grouping adjustable loads, the grouped active power and total active power of each adjustable load group are obtained, as shown in the following formula:

[0109] (11)

[0110] In the formula, For the first The sum of the active power of the load groups; For the first The first in the group load Active power of each load; This represents the total active power of all adjustable loads.

[0111] The calculation methods for the adjustable capacity of each load group and the total adjustable capacity are as follows:

[0112] (12)

[0113] In the formula, For the first Adjustable capacity of the load group; For the first The first in the group load The coefficient of active power of each load; The total adjustable capacity of the adjustable load; For the first Adjustable coefficient of active power of group load; This is the total adjustability coefficient for adjustable loads.

[0114] It is worth noting that the active power and total power of each adjustable load group are used to monitor the operation of the adjustable loads; the adjustable capacity of each adjustable load group and the total adjustable capacity are used for actual adjustment and control. This is similar to controlling power changes when adjusting temperature in an air conditioner, but it requires monitoring the actual power.

[0115] S6. Store the adjustment status of all the above data and adjustment commands in real time.

[0116] As shown in Table 1, the adjustment status of all the above real-time data, calculated values, and issued control commands is stored in real time. The main storage structure of the relational database is as follows:

[0117] (1) The real-time data storage fields in the database include: ① Telemetry, measurement point name, measurement point ID, measurement point value, measurement point refresh time, previous value, whether refreshed, whether communication is normal, whether it has changed, whether it has changed abruptly, whether it has exceeded the limit, the value of exceeding the limit, historical storage period, etc.; ① Remote signaling, switch / disconnector name, switch / disconnector status, open position, closed position, refresh time, whether refreshed, whether communication is normal, change storage, etc.

[0118] (2) The demand calculation storage fields in the database include: demand metering point name, demand metering point ID, demand metering point demand limit (maximum demand value), demand time period, metering point real-time active power, metering point demand time period, metering point demand future time period, metering point demand electricity, metering point demand future electricity limit value, metering point demand future power limit value, etc.

[0119] (3) The fields for issuing control instructions in the database include: load name, load ID, load corresponding measurement point, load real-time value, load adjustable value, load adjustment rate, load adjustable capacity, load level, whether the load can be interrupted, load adjustment value, load future time period predicted average value, load adjustment completion time, etc.

[0120] Table 1 List of Adjustable Loads

[0121]

[0122] S7. Using the K-means clustering algorithm of artificial intelligence, the adjustable load within the enterprise is divided into k clusters based on the stored data to achieve load grouping, and multiple time periods are divided for optimization.

[0123] The specific implementation process of step S7 is as follows:

[0124] S7.1 Preprocess the stored data, including removing abnormal data such as non-refreshed data and communication interruptions, and manually adjusting the load importance level.

[0125] S7.2. Select an appropriate value of k using the Elbow Method, and assign each adjustable load to the nearest group center to form k groups (clustering).

[0126] S7.3 Run the k-means algorithm to recalculate the center of each group (i.e., the mean of all points in each group).

[0127] S7.4 Determine whether the clustering has converged or the number of iterations has been reached. If not, repeat steps S7.2 and S7.3 until the group centers no longer change or the preset number of iterations is reached, and output the clustering results.

[0128] S7.5 Analyze the calculation results and view the characteristics of different groups, such as peak load periods, load types, and load levels.

[0129] S7.6 The artificial intelligence clustering K-means algorithm runs in real time and feeds back the grouping results to the demand analysis and control algorithm in real time.

[0130] The K-means clustering algorithm based on artificial intelligence is used to divide the adjustable load within the enterprise into k clusters to achieve load grouping. Data points are assigned according to the principle of nearest distance and the center position is updated until the termination condition is met.

[0131] S8. After the end of the current month's demand monitoring cycle, the maximum demand for the current month will be calculated based on historical data. This will serve as the basis for enterprises to sign demand limit agreements with the power grid in the next month, thereby reducing the enterprises' demand limit.

[0132] Operators use workstation monitors to monitor the calculation and control results of the demand monitoring system in real time, and update the demand limits of the demand metering points according to demand.

[0133] Operators can monitor the power demand monitoring system in real time through the workstation monitor. The system allows for convenient, fast, and safe monitoring of changes in the company's power demand through real-time images and trend curves, a power demand control human-machine interface, and an adjustable load priority editing human-machine interface.

[0134] This invention primarily addresses the issue of reducing electricity costs for enterprises based on demand-based pricing. The system and method are deployed on the enterprise's centralized power control system. By acquiring real-time telemetry data and switch status of voltage, current, active power, and other parameters of the maximum demand metering point and various loads within the enterprise, it calculates the demand value at different time intervals from the maximum demand metering point. Using the equal-area and step-by-step approximation methods, combined with the enterprise's load importance classification, it calculates the loads requiring regulation and issues control commands, achieving precise monitoring of the enterprise's electricity demand and dynamic load management. This invention innovatively introduces the equal-area and step-by-step approximation methods to establish an enterprise demand control model based on maximum demand limits. This method provides enterprises with an economical and reliable electricity demand management solution, and has significant application value in promoting energy conservation, emission reduction, cost reduction, and efficiency improvement for enterprises.

[0135] This invention overcomes the problems and defects of existing technologies, such as the inability to transfer loads, repetitive demand calculations, the inability to establish a plant-level adjustable load model, and large errors in the prediction algorithms. It provides a method and system for monitoring the electricity demand of industrial enterprises. The invention proposes using the rolling slip method for demand calculation, the equal area method for power limitation, the multi-time-period stepwise approximation method for demand control, and the K-means algorithm for adjustable load grouping. This enables refined monitoring of enterprise electricity demand, improves the work efficiency of enterprise staff, reduces the enterprise's demand-related electricity costs, and enhances the enterprise's competitiveness.

[0136] Example 2

[0137] like Figure 5 As shown, an industrial enterprise power demand monitoring system of the present invention includes:

[0138] The power centralized control data acquisition and processing module is used to acquire the power data of the load through the enterprise's power centralized control system, and pre-group the enterprise's adjustable loads according to the adjustment priority based on the power data;

[0139] The electricity demand calculation and analysis module is used to calculate the demand time period using the rolling slip calculation method. The demand within the period is calculated on a rolling basis using a slip time scale, where the slip time scale and the demand period are... Adjust according to demand;

[0140] The electricity demand calculation and analysis module is used to calculate the demand time period. The calculation is divided into two time periods: the period that has already occurred and the period that will occur in the future. The rolling slip calculation method is used to calculate the electricity consumption during the period that has already occurred. and demand time period Battery During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. Average active power over the future time period Less than the average active power limit for the future time period Proceed to the next step if the conditions are met;

[0141] The electricity demand calculation and analysis module is used to divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the demand for electricity. The group has experienced a period of time in terms of electricity consumption. Combined with the electricity demand metering point's electricity limit and the The group has experienced a period of time in terms of electricity consumption. Calculate the first Average active power limit for future time period The stepwise approximation method is adopted, based on the first... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in stages, in the first stage... Average active power of the group in the future time period Less than In this case, further determine the real-time active power at the demand metering point. Is it less than ;

[0142] The power demand control command issuing module is used to control the real-time active power at the demand metering point. Exceed Based on the pre-grouping of adjustable loads, the grouped active power of each adjustable load and the total active power of all adjustable loads are obtained; combined with the grouped active power of each adjustable load, the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. Based on the adjustable capacity of the j-th load group Total adjustable capacity of adjustable load Adjustment commands are issued to the adjustable load in groups and time periods for adjustment;

[0143] The power demand control command issuing module is used to store all the above data and the adjustment status of the adjustment commands in real time.

[0144] The AI-based K-means clustering algorithm module is used to divide the adjustable load within an enterprise into k clusters based on stored data, thereby grouping the adjustable load. After grouping, it determines the real-time power of the demand metering points after adjustment. If the active power is less than the average active power limit for multiple future time periods, then increase the regulating load and readjust.

[0145] The data statistics module is used to calculate the maximum demand for the month based on historical data after the end of the monthly demand monitoring cycle, which serves as the basis for enterprises to sign demand limit agreements with the power grid in the following month.

[0146] In one optional implementation, the industrial enterprise power demand monitoring method includes: a) acquiring load power data through the enterprise's power centralized control system, and pre-grouping the enterprise's adjustable loads according to adjustment priority based on the power data; b) continuously calculating the demand within a demand time period using a rolling slip calculation method; c) calculating the power limit at the power demand metering point, and further calculating the average active power limit for future time periods; d) dividing the demand time period into... The load is divided into multiple groups, and the average active power limit for future time periods is calculated for each group. The adjustable loads in the pre-grouped groups are then adjusted in a step-by-step manner. e) The adjustable capacity and total adjustable capacity of each load group are calculated, and adjustment instructions are issued to the adjustable loads in groups and time periods. f) All the above data and the adjustment status of the adjustment instructions are stored in real time. g) The K-means clustering algorithm is used to divide the adjustable loads within the enterprise into k clusters based on the stored data, and the division into multiple time periods is optimized. h) After the demand monitoring cycle ends this month, the maximum demand value for this month is calculated based on historical data, which serves as the basis for the enterprise to sign the demand limit agreement with the power grid for the next month.

[0147] In addition to the modules mentioned above, the industrial enterprise power demand monitoring system of the present invention also includes server 1a, server 1b, workstation 2, and switch 3. Switch 3 comprises four switches: two of which serve as internal communication switches for servers 1a, 1b, and workstation 2; the other network ports of each server are connected to the other two switches for data exchange and command issuance, all in a dual-network redundant configuration.

[0148] Install operating system 4 on server 1a, server 1b and workstation 2. Simultaneously, install and deploy relational database 5 on operating system 4 on server 1a and server 1b, and install and deploy real-time database 6 on operating system 4 on server 1a, server 1b and workstation 2.

[0149] Relational database 5 on servers 1a and 1b is synchronized using cluster software 14. Real-time database 6 on servers 1a, 1b, and workstation 2 is synchronized using real-time database synchronization software 15.

[0150] Servers 1a and 1b deploy a power centralized control data acquisition and processing module 8, a power demand calculation and analysis module 9, a power demand control command issuance module 10, and an artificial intelligence clustering K-means algorithm module 13. The server backend service implements the above power demand calculation, load grouping, demand time period segmentation, multiple groups of future time period limits, load adjustment value allocation, and command issuance.

[0151] Workstation 2 is equipped with a real-time screen and trend curve display module 7, a power demand control human-machine interface 11, and an adjustable load priority editing human-machine interface 12. The power demand control human-machine interface 11 is deployed and runs on the workstation for convenient monitoring and operation. Operators can conveniently and quickly monitor changes in the company's power demand through the real-time screen and trend curves of the power demand monitoring system, the power demand control human-machine interface, and the adjustable load priority editing human-machine interface.

[0152] Specifically, the background processing module and services run simultaneously on servers 1a and 1b, forming a hot standby mode. The failure of any one server or its module will not affect the execution of the entire program. The human-machine interface and operations are performed on workstation 2, facilitating monitoring by staff.

Claims

1. A method for monitoring the electricity demand of industrial enterprises, characterized in that, include: The power data of the load is obtained through the enterprise's power centralized control system, and the enterprise's adjustable loads are pre-grouped according to the adjustment priority based on the power data; The demand time period is calculated using the rolling slip calculation method. The demand within the period is calculated on a rolling basis using a slip time scale, where the slip time scale and the demand period are... Adjustments are made according to demand; the rolling slip calculation method is that the power consumption in the middle time period remains unchanged, the slip period at the very beginning of the demand time period is automatically rolled out, and the power consumption in the latest slip period period is automatically rolled in to obtain the power consumption in the new demand time period. Demand time period Dividing the data into two time periods, already occurred and in the future, the rolling slip calculation method is used to calculate the electricity consumption during the already occurred time period. and demand time period Battery During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. Average active power over the future time period Less than the average active power limit for the future time period Proceed to the next step if the conditions are met; Divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the first... The group has experienced a period of time in terms of electricity consumption. Combined with the electricity demand metering point's electricity limit and the The group has experienced a period of time in terms of electricity consumption. Calculate the first Average active power limit for future time period ; Using the stepwise approximation method, based on the first... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in stages, in the first stage... Average active power of the group in the future time period Less than In this case, further determine the real-time active power at the demand metering point. Is it less than ; If the demand metering point has real-time active power Exceed Based on the pre-grouping of adjustable loads, the grouped active power of each adjustable load and the total active power of all adjustable loads are obtained; combined with the grouped active power of each adjustable load, the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. According to the first Adjustable capacity of load group Total adjustable capacity of adjustable load Adjustment commands are issued to the adjustable load in groups and time periods for adjustment; Real-time storage of load power data, pre-grouping of adjustable loads according to adjustment priority, power demand calculation data, and adjustment status of adjustment commands; The K-means clustering algorithm, based on artificial intelligence, is used to divide the adjustable load within the enterprise into k clusters according to the stored data, thereby grouping the adjustable load. After grouping, the real-time active power at the demand metering point after adjustment is determined. If the active power is less than the average active power limit for multiple future time periods, then increase the regulating load and readjust. After the demand monitoring period ends this month, the maximum demand for this month will be calculated based on historical data, which will serve as the basis for enterprises to sign demand limit agreements with the power grid next month.

2. The method for monitoring the electricity demand of industrial enterprises according to claim 1, characterized in that, The method employs a rolling slip calculation to continuously calculate the demand over a time period, using a slip time scale for rolling calculations. The slip time scale and the demand time period are... Adjustments will be made according to demand, including: Assuming the demand period is 15 minutes, starting from the first minute of the month, the active power is integrated from minute 0 to minute 15 to obtain the electricity consumption for the first 15-minute period. The average active power for this period is obtained by dividing the electricity consumption for this 15-minute period by the time period. When the time reaches the 16th minute, the electricity consumption from minute 0 to 1 is automatically subtracted, and the electricity consumption from minute 15 to 16 is automatically added to the total electricity consumption for the new 15-minute period. The average active power for the new time period is also calculated, as shown in the following formula: ； In the formula, From Battery level for the next 15 minutes from this time; From Battery level for one minute from the current time; for Battery level one minute prior to the event; From The average active power over the next 15 minutes from that time; Real-time active power at the demand metering point; From The average active power over the next 15 minutes from that time; The start time for demand calculation; This is the end time for demand calculation.

3. The method for monitoring the electricity demand of industrial enterprises according to claim 1, characterized in that, The required time period Dividing the data into two time periods, already occurred and in the future, the rolling slip calculation method is used to calculate the electricity consumption during the already occurred time period. and demand time period Battery ,include: Demand time period Divided into two time periods: already occurred and future. , Using the equal area method, calculate the electricity values ​​for the past and future time periods: ； In the formula, This represents the amount of electricity consumed during the time period in question. This is the current calculation time; This refers to the start time of a past time period. For future battery life; This is the end time of a future time period; Real-time active power at the demand metering point; Demand period Battery level.

4. The method for monitoring the electricity demand of industrial enterprises according to claim 1, characterized in that, The electricity limit combined with the electricity demand metering point Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. ,include: Electricity demand limit is the demand time period Given the maximum average value of internal active power, the expression for the electricity limit at the electricity demand metering point is as follows: ； In the formula, Demand limits for electricity demand metering points; The electricity consumption limit for electricity demand metering points; For the time period of demand; Combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. The expression is as follows: 。 5. The method for monitoring the electricity demand of industrial enterprises according to claim 1, characterized in that, The process involves dividing the past and future time periods into multiple groups, with each group including both past and future time periods. The electricity consumption during the i-th group of past time periods is then calculated. ,include: Demand time period Divided into multiple groups, each representing a past event and a future event, respectively. , ... ,in, For the first Group; using the method of equal area of ​​electric charge, calculate the first group; The electricity consumption values ​​for the two time periods are shown in the following formula: ； In the formula, For the first Group demand for electricity during a specific time period; For the first The group has consumed electricity over the specified period of time; This is the current calculation time; For the first The start time of the past time period; For the first Group the amount of electricity to be generated in the future time period. For the first The end time of the future time period.

6. The method for monitoring the electricity demand of industrial enterprises according to claim 5, characterized in that, The first The expression for the average active power limit for the future time period is as follows: ； In the formula, For the first Average active power limit for the group over a future time period; The electricity consumption limit for electricity demand metering points; For the first The group has consumed electricity over the specified period of time; For the first Group of future time periods.

7. The method for monitoring the electricity demand of industrial enterprises according to claim 1, characterized in that, The calculation methods for the grouped active power of each adjustable load and the total active power of all adjustable loads are as follows: ； In the formula, For the first The sum of the active power of the load groups; For the first The first in the group load Active power of each load; This represents the total active power of the load.

8. The method for monitoring the electricity demand of industrial enterprises according to claim 7, characterized in that, The active power of each adjustable load group is calculated by combining the grouped active power. Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. The expression is as follows: ； In the formula, For the first Adjustable capacity of the load group; For the first The first in the group load The coefficient of active power of each load; The total adjustable capacity of the adjustable load; For the first Adjustable coefficient of active power of group load; This is the total adjustability coefficient for adjustable loads.

9. The method for monitoring the electricity demand of industrial enterprises according to claim 1, characterized in that, The method employs the K-means clustering algorithm based on artificial intelligence. Based on stored data, it divides the adjustable load within the enterprise into k clusters, thereby grouping the adjustable load, including: The stored data is preprocessed, including the removal of outlier data and the human intervention of load importance levels; By selecting an appropriate value of k using the Elbow Method, each adjustable load is assigned to the nearest group center, forming k groups, thus achieving the grouping of adjustable loads. Run the k-means algorithm to recalculate the center of each group, that is, the mean of all points in each group; Determine whether the clustering has converged or reached the desired number of iterations. If not, repeat the above steps until the group centers no longer change or the preset number of iterations is reached, and then output the clustering results. Analyze the calculation results and examine the characteristics of different groups; The K-means clustering algorithm for artificial intelligence runs in real time and feeds back the grouping results to the demand analysis and control algorithm in real time.

10. An industrial enterprise power demand monitoring system, characterized in that, include: The power centralized control data acquisition and processing module is used to acquire the power data of the load through the enterprise's power centralized control system, and pre-group the enterprise's adjustable loads according to the adjustment priority based on the power data; The electricity demand calculation and analysis module is used to calculate the demand time period using the rolling slip calculation method. The demand within the period is calculated on a rolling basis using a slip time scale, where the slip time scale and the demand period are... Adjustments are made according to demand; the rolling slip calculation method is that the power consumption in the middle time period remains unchanged, the slip period at the very beginning of the demand time period is automatically rolled out, and the power consumption in the latest slip period period is automatically rolled in to obtain the power consumption in the new demand time period. The electricity demand calculation and analysis module is used to calculate the demand time period. Dividing the data into two time periods, already occurred and in the future, the rolling slip calculation method is used to calculate the electricity consumption during the already occurred time period. and demand time period Battery During the demand period Battery Electricity consumption less than the electricity demand metering point limit In this case, combined with the electricity demand metering point's electricity limit Electricity consumption during the period in which the event occurred The average active power limit for the future time period is calculated. Average active power over the future time period Less than the average active power limit for the future time period Proceed to the next step if the conditions are met; The electricity demand calculation and analysis module is used to divide the past and future time periods into multiple groups, each group including both past and future time periods, and calculate the demand for electricity. The group has experienced a period of time in terms of electricity consumption. Combined with the electricity demand metering point's electricity limit and the The group has experienced a period of time in terms of electricity consumption. Calculate the first Average active power limit for future time period ; Using the stepwise approximation method, based on the first... Average active power limit for future time period The adjustable load of pre-grouped enterprises is gradually adjusted in a step-by-step manner, with the average active power of the i-th group in the future time period. Less than In this case, further determine the real-time active power at the demand metering point. Is it less than ; The power demand control command issuing module is used to control the real-time active power at the demand metering point. Exceed Based on the pre-grouping of adjustable loads, the grouped active power of each adjustable load and the total active power of all adjustable loads are obtained; combined with the grouped active power of each adjustable load, the first... Adjustable capacity of load group The total adjustable capacity of the adjustable load is calculated by combining the total active power of all adjustable loads. According to the first Adjustable capacity of load group Total adjustable capacity of adjustable load Adjustment commands are issued to the adjustable load in groups and time periods for adjustment; The power demand control command issuing module is used to store the load's power data, the pre-grouping of adjustable loads according to adjustment priority, the power demand calculation data, and the adjustment status of the adjustment commands in real time. The AI-based K-means clustering algorithm module is used to divide the adjustable load within an enterprise into k clusters based on stored data, thereby grouping the adjustable load. After grouping, it determines the real-time active power of the demand metering points after adjustment. If the active power is less than the average active power limit for multiple future time periods, then increase the regulating load and readjust. The data statistics module is used to calculate the maximum demand for the month based on historical data after the end of the monthly demand monitoring cycle, which serves as the basis for enterprises to sign demand limit agreements with the power grid in the following month.

Images