Power consumption management system and method in power load

Abstract

The invention belongs to the field of power consumption management, and particularly relates to a power consumption management system and method in a power load, and the system consists of a data module, a load module, an intelligent switch module, a multi-objective optimization module and a load prediction module, the data module is in butt joint with the load module through the intelligent switch module, and the load prediction module is in butt joint with the load module through the multi-objective optimization module. The intelligent switch module is used for acquiring power load data of electric equipment and transmitting the power load data to the load prediction module, the load prediction module transmits current day load prediction data to the multi-target optimization control module, and the multi-target optimization control module connects an output state with the intelligent switch module and transmits an optimization result to the intelligent switch module. According to the method, the loads are classified, and the equipment is monitored and managed, so that the equipment reasonably runs to reach the running state within the load range.

Description

Technical Field

[0001] The present invention belongs to the field of electricity management, and in particular relates to an electricity management system and method within an electric load. Background Art

[0002] As the contradiction between energy supply and consumption becomes increasingly serious, many power suppliers have adopted a strategy of using different rates for peak and valley periods to balance the power load and save energy. Therefore, many users hope to manage the load of their own power equipment, for example, to optimize the operation schedule of the power equipment to avoid running the equipment during peak hours as much as possible, thereby saving electricity expenses. Or the supplier configures the power system of the area according to the number of electricity users in the power consumption area, so as to ensure the stable and efficient use of electricity in each area. However, in the prior art, the energy management system for load management is generally used in large-scale manufacturing enterprises with high energy consumption, and the relevant data is aggregated on a powerful processing platform, and then load management is performed. This process is not suitable for ordinary users to supply electricity, and complex technology is required to realize the function of optimizing the operation schedule in load management. In addition, the population using electricity in each region has a large mobility, so it is not a good way to judge the actual population of electricity users in each region every day. Therefore, in the supply of electricity, it is difficult to manage power distribution in combination with the actual population.

[0003] Therefore, the present invention provides an electricity management system and method within the power load, which classifies the loads, monitors the equipment, and manages it to operate reasonably to achieve an operating state within the load range. Summary of the invention

[0004] The present invention provides an electricity management system and method within an electric load, which is used to solve the technical problem that the prior art lacks a reasonable load optimization configuration scheme for electric loads, which is not conducive to the safe and stable operation of the power grid.

[0005] In order to achieve the above-mentioned purpose, the technical scheme adopted by the present invention is as follows: an electricity management system within the power load, consisting of a data module, a load module, an intelligent switch module, a multi-objective optimization module and a load prediction module. The data module is connected to the load module through the intelligent switch module to obtain the power load data of the power-consuming equipment, and transmits the power load data to the load prediction module. The load prediction module transmits the load prediction data of the day to the multi-objective optimization control block. The multi-objective optimization control block connects the output state to the intelligent switch module and transmits the optimization result to the intelligent switch module, thereby controlling the load module to maintain normal operation of different loads, thereby ensuring that the electric energy load is within the allowable range. The multi-objective optimization module is optimized according to five optimization conditions, and the optimization objective function includes two constraints: temperature and power.

[0006] Furthermore, the data module includes a data acquisition module, a data transmission module and a data storage and processing module. The data acquisition module collects power load data in real time through sensors and smart meters installed at each power load point, including voltage, current, power factor, active power, reactive power and power output data. The data acquisition module will upload the power load data to the data storage and processing module for processing through the data transmission module.

[0007] Furthermore, the data storage and processing module improves the quality and availability of the data by cleaning, supplementing missing values, standardizing, normalizing and filling missing values ​​of the collected power load data.

[0008] Further, the load module includes a rigid load, an interruptible load and an uninterruptible load.

[0009] Furthermore, the intelligent switch module is connected to the data module and the load prediction module.

[0010] Furthermore, the power load data processed by the data storage and processing module is connected to the load prediction module through the intelligent switch module interface 1, and the intelligent switch module interface 2 is connected to the multi-objective optimization control block. The intelligent switch module interface 2 divides the input layer into two categories: the state within the load range and the state outside the load range. When the state outside the load range is detected, the sensor alarm circuit in the intelligent switch module interface 2 triggers a signal to provide early warning and suggestions.

[0011] Furthermore, the load prediction module analyzes historical power load data by improving the LiBSVM, i.e., C-LiBSVM model, predicts changes in grid load and power output in the future, establishes a C-LiBSVM prediction model between voltage data, state data and grid load, and establishes a C-LiBSVM prediction model between current data, state data and power output.

[0012] Furthermore, the optimization conditions include user experience comfort, carbon emissions, daily load, time-of-use electricity price, and user equipment usage intention.

[0013] Furthermore, the multi-objective optimization module adopts an optimization model of the load power management system based on time-of-use electricity prices and multi-objective optimization. The model classifies the loads and monitors the equipment, and achieves reasonable operation and operating status within the load range through an improved particle swarm algorithm.

[0014] A method for managing power consumption in an electrical load comprises the following steps:

[0015] (1) Data collection: The data collection module collects power load data in real time, including voltage, current, power factor, active power, reactive power and power output data, through sensors and smart meters installed at each power load point. The data collection module uploads the power load data to the data storage and processing module for processing through the data transmission module;

[0016] (2) Data processing: The data storage and processing module cleans, supplements missing values, standardizes, normalizes, and fills missing values ​​on the collected power load data, thereby improving the quality and availability of the data;

[0017] (3) Load optimization: The load prediction module transmits the load prediction data of the day to the multi-objective optimization control block. The multi-objective optimization control block connects the output state to the intelligent switch module and transmits the optimization result to the intelligent switch module, thereby controlling the load module to maintain the normal operation of different loads, thereby ensuring that the power load is within the allowable range.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] (1) This invention considers multi-dimensional optimization objectives - user comfort, electricity price cost and carbon emission tracking - to establish a multi-objective optimization model under time-of-use electricity price to ensure optimal scheduling under different load power consumption behaviors. An optimization model for load power management system based on time-of-use electricity price and multi-objective optimization is proposed. The model classifies the loads, monitors the equipment, and manages it to achieve reasonable operation within the load range.

[0020] (2) The present invention combines a prediction algorithm with an analysis of historical data to enable the system to predict changes in grid load and power output over a period of time in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a structural diagram of the power management system within the power load of the present invention;

[0022] Figure 2 This is a circuit diagram of the control circuit of the present invention. DETAILED DESCRIPTION

[0023] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0024] like Figure 1As shown, a power management system within the power load is composed of a data module, a load module, an intelligent switch module, a multi-objective optimization module and a load prediction module. The data module is connected to the load module through the intelligent switch module to obtain the power load data of the power equipment, and transmits the power load data to the load prediction module. The load prediction module transmits the load prediction data of the day to the multi-objective optimization control block. The multi-objective optimization control block connects the output state to the intelligent switch module and transmits the optimization result to the intelligent switch module, thereby controlling the load module to keep different loads running normally, thereby ensuring that the power load is within the allowable range. The multi-objective optimization module is optimized according to five optimization conditions, and the optimization objective function includes two constraints: temperature and power.

[0025] The data module includes a data acquisition module, a data transmission module and a data storage and processing module. The data acquisition module collects power load data in real time through sensors and smart meters installed at each power load point, including voltage, current, power factor, active power, reactive power and power output data. The data acquisition module will upload the power load data to the data storage and processing module for processing through the data transmission module.

[0026] The data storage and processing module improves the quality and availability of the data by cleaning, supplementing missing values, standardizing, normalizing and filling missing values ​​on the collected power load data.

[0027] Data cleaning: Use the standard deviation method to identify and handle outliers. For each data point x i (Historical power load data i , voltage U i 、Current I i , Active power P i , power factor μ i , power output data M i

[0028] x i =ω u ·u i +ω U ·U i +ω I I i +ω P ·P i +ω μ ·μ i +ω M ·M i , i∈{1,2,3...,N},), whose

[0029] Medium xx is the normalized weight value. Calculate its average value The difference χ i =x i -u, and with standard deviation If the difference χ i If the value of -s exceeds the threshold m=3s (3 times the standard deviation), the data point is considered an outlier and is removed or replaced.

[0030] Data standardization: Standardize data from different sources and units so that they can be compared and analyzed on the same scale.

[0031] Data standardization method: Improved Z-score standardization,

[0032] Among them, x is the relevant power load data after data cleaning, μ is the mean of the data, and σ is the standard deviation of the data. After standardization, the mean of the data is 0 and the standard deviation is 1.

[0033] Data normalization: Convert data into dimensionless values ​​between 0 and 1 to facilitate subsequent algorithm processing and analysis.

[0034] Minimum-maximum normalization formula:

[0035] Among them, x is the relevant power load data after data cleaning, x min and x max are the minimum and maximum values ​​in the data respectively. After normalization, the value range of the data is between 0 and 1.

[0036] Missing value filling: Using linear interpolation, for missing values ​​x m , whose two non-missing values ​​are x a and x b , and a<m<b, then

[0037] The load module includes rigid load, interruptible load and non-interruptible load. Rigid load means that the use time of electrical equipment will not change according to the optimization system, such as refrigerators, electric lights, etc. Interruptible loads are divided into two categories according to whether the equipment use time can be interrupted during the translation process: ① Translable non-interruptible loads, such as dishwashers and washing machines, whose operating time can be moved and adjusted but the equipment cannot be interrupted during operation; ② Translable and interruptible loads, which means that the use time of such household equipment changes according to the optimization system and can be interrupted during operation, such as electric cars and water heaters.

[0038] The intelligent switch module is connected to the data block and the load prediction module. First, the data module processes the relevant power load data x iThe intelligent switch module interface 1 is connected to the load prediction module. The intelligent switch module interface 2 is connected to the multi-objective optimization control block. The intelligent switch module interface 2 divides the input layer into two categories: the state within the load range and the state outside the load range. When the state outside the load range is detected, a trigger signal is given to the sensor alarm circuit in the intelligent switch module interface to provide corresponding warnings and suggestions. Alarm circuit Figure 2 As shown:

[0039] First, the state signal of the input layer is encoded through the encoder and connected to relay1. When the Realay1 interface inputs a low level, the transistor is in the off state and the relay does not work, so that the alarm circuit does not work. When connected to a high level, after the voltage is divided by the current-limiting resistor R14, the transistor is in the on state. At this time, the induction coil in the relay passes current, generating a magnetic field so that Vout outputs a 12V voltage and the excitation circuit works. The alarm is completed, and the indicators to be optimized are given, which are fed back to the multi-objective optimization control module to continue optimization until it is within the load range.

[0040] The present invention predicts the changes of power grid load and power output within a future period of time by analyzing historical data.

[0041] Based on these prediction results, the system can formulate voltage control strategies in advance to achieve preventive control. This predictive control can adjust control parameters in advance before the grid load or power output changes significantly, thereby avoiding risks such as voltage over-limit or equipment damage.

[0042] Prediction algorithm: The improved LiBSVM (C-LiBSVM) model is used for prediction.

[0043] Model 1: Establishing a C-LiBSVM prediction model between voltage data, state data and “grid load”

[0044] Model 2: Establishing a C-LiBSVM prediction model between current data, state data and “power output”

[0045] Variable: X1-X6 (current I i , voltage U i , power factor μ i , load data u i , power output data M i , active power P i )

[0046] Model 1 Explained variables: Y1 (grid load)

[0047] Model 2 Explained variable: Y2 (power output)

[0048] The prediction effect of the C-LiBSVM prediction model can be evaluated by the mean absolute percentage error (MAPE) and the mean absolute error (MAE). The expressions of MAPE and MAE are as follows:

[0049]

[0050] In order to speed up the prediction, the above variables are normalized: the eigenvalues ​​are mapped from a large range to [-1,1];

[0051] f(x):x→y,y∈[-1,1]

[0052]

[0053] Multi-objective optimization control block: Considering multi-dimensional optimization objectives - user comfort, electricity price cost and carbon emission tracking - a multi-objective optimization model under time-of-use electricity price is established to ensure optimal scheduling under different load power consumption behaviors. An optimization model for load power management system based on time-of-use electricity price and multi-objective optimization is proposed. The model classifies the load, monitors the equipment, and manages it to operate reasonably to achieve the operating state within the load range.

[0054] The modeling is as follows: Power load model

[0055] (1) Rigid load

[0056] P c =S c (t)*P c,rated

[0057] E c =P c,rated *d c

[0058]

[0059] Where: c∈C is the rigid load; P c is the instantaneous power of load c at time t; S c (t) is the working state S of load c at time t c (t) = 1 means the device is in the on state, S c (t) = 0 means the load is in the off state; P c，rated is the rated power of load c is the load on time, is the load end time; E c is the total power consumption of load C; d c is the total working time of the rigid load C.

[0060] (2) Non-interruptible load

[0061] T i start,s ≤T i start ≤T i start,e

[0062] T i end =T i start +d i

[0063] E i =p i,rated +d i

[0064] Where: i∈I is the non-interruptible load; T i start,s 、T i start,e is the upper and lower limits of the start time range; T i start is the start-up time of load i; T i end is the end time of load i; d i , is the working time of load i; E i is the total power consumption of the non-interruptible load i.

[0065] (3) Interruptible load

[0066] T k start,s ≤T k,1 start ≤T k start,e

[0067]

[0068] E k =P k,start *d k

[0069] Where: k∈K is the interruptible load; is the start time of the jth working period of load k; is the end time of the jth working period of load k; M is the total number of working periods. k is the total power consumption of the interruptible loads.

[0070] Optimization objective function: The economic objective function takes household electricity costs as the optimization target.

[0071] The smaller the optimized value is, the better the optimization effect is.

[0072]

[0073] Where: F1 is the electricity cost used up to time t; P a (t) is the electricity consumption of the equipment in period t; i(t) is the time-of-use electricity price of the equipment in period t.

[0074] User experience comfort function

[0075]

[0076] Where: x is the user satisfaction factor for load x, β x The value is a non-negative integer. The larger the impact factor, the more likely the user wants this load to be completed first within the optimization time. x The user's comfort level with the electrical load x; The time that the electrical load x needs to work; The start / stop time of the power load x that the user wants; is the actual stop time of the power load x; F2 is the total power load satisfaction; K is the total number of dispatchable loads.

[0077] Carbon emissions

[0078]

[0079] Where: F3 is carbon emissions; P a (t) is the power consumed by load a.

[0080] Constraints

[0081] Power Constraints

[0082]

[0083] Where: W i Indicates the average power of the load; X ij Indicates the working state of the i-th load in the j-th time period. The value is 1 when it is working and 0 when it is not working. j Indicates the total power of the loads that cannot be dispatched or must be running; P j It represents the power provided by the grid during time period j.

[0084] Temperature constraints

[0085] T K (t) = T s (t)+ΔT

[0086] Where: T k (t) is the actual temperature of the room at time t; Ts (t) is the ideal room temperature at time t; △T is the temperature difference allowed by the user.

[0087] Improved particle swarm optimization algorithm

[0088]

[0089] Where: x s,d is the position of the sth particle in the d-dimensional search space; v s,d is the velocity of the particle; p s,d is the individual extreme value; p g,d is the population extreme value; w is the inertia factor; c1 and c2 are non-negative constants, called learning factors; r1 and r2 are random numbers distributed between 0 and 1; k is the current iteration number; k max is the maximum number of iterations.

[0090]

[0091] Where: f is the current fitness value; is the average fitness value; f min is the minimum fitness value; w max 、w min are the maximum and minimum values ​​of the inertia factor respectively.

[0092] The multi-objective optimization module adopts an optimization model of the load power management system based on time-of-use electricity prices and multi-objective optimization. The model classifies the loads and monitors the equipment. It achieves reasonable operation and reaches the operating state within the load range through the improved particle swarm algorithm.

[0093] A method for managing power consumption in an electrical load comprises the following steps:

[0094] (1) Data collection: The data collection module collects power load data in real time, including voltage, current, power factor, active power, reactive power and power output data, through sensors and smart meters installed at each power load point. The data collection module uploads the power load data to the data storage and processing module for processing through the data transmission module;

[0095] (2) Data processing: The data storage and processing module cleans, supplements missing values, standardizes, normalizes, and fills missing values ​​on the collected power load data, thereby improving the quality and availability of the data;

[0096] (3) Load optimization: The load prediction module transmits the load prediction data of the day to the multi-objective optimization control block, which connects the output state to the intelligent switch module and transmits the optimization result to the intelligent switch module, thereby controlling the load module to keep different loads running normally, thereby ensuring that the electric energy load is within the allowable range.

[0097] The present invention considers multi-dimensional optimization objectives - user comfort, electricity price cost and carbon emission tracking - to establish a multi-objective optimization model under time-of-use electricity price, ensuring the optimal scheduling under different load power consumption behaviors. An optimization model for load power management system based on time-of-use electricity price and multi-objective optimization is proposed. The model classifies the load, monitors the equipment, and manages it to achieve reasonable operation within the load range.

[0098] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A power management system within a power load, characterized in that: It consists of a data module, a load module, an intelligent switch module, a multi-objective optimization module and a load prediction module. The data module is connected to the load module through the intelligent switch module to obtain the power load data of the electrical equipment and transmit the power load data to the load prediction module. The load prediction module transmits the load prediction data of the day to the multi-objective optimization control block. The multi-objective optimization control block connects the output state to the intelligent switch module and transmits the optimization result to the intelligent switch module, thereby controlling the load module to keep different loads running normally, thereby ensuring that the electric energy load is within the allowable range. The multi-objective optimization module is optimized according to five optimization conditions, and the optimization objective function includes two constraints: temperature and power.

2. A power management system within a power load according to claim 1, characterized in that: The data module includes a data acquisition module, a data transmission module and a data storage and processing module. The data acquisition module collects power load data in real time through sensors and smart meters installed at each power load point, including voltage, current, power factor, active power, reactive power and power output data. The data acquisition module will upload the power load data to the data storage and processing module for processing through the data transmission module.

3. A power management system within a power load according to claim 2, characterized in that: The data storage and processing module improves the quality and availability of the data by cleaning, supplementing missing values, standardizing, normalizing and filling missing values ​​on the collected power load data.

4. The power management system in the power load according to claim 1, characterized in that: The load modules include rigid loads, interruptible loads and non-interruptible loads.

5. The power management system in the power load according to claim 1, characterized in that: The intelligent switch module is connected with the data module and the load prediction module.

6. The power management system in the power load according to claim 4, characterized in that: The power load data processed by the data storage and processing module is connected to the load prediction module through the intelligent switch module interface 1, and the intelligent switch module interface 2 is connected to the multi-objective optimization control block. The intelligent switch module interface 2 divides the input layer into two categories: the state within the load range and the state outside the load range. When the state outside the load range is detected, the sensor alarm circuit in the intelligent switch module interface 2 triggers a signal to provide early warning and suggestions.

7. The power management system in the power load according to claim 1, characterized in that: The load prediction module analyzes historical power load data through an improved LiBSVM, namely, a C-LiBSVM model, predicts changes in grid load and power output in the future, establishes a C-LiBSVM prediction model between voltage data, state data and grid load, and establishes a C-LiBSVM prediction model between current data, state data and power output.

8. The power management system in the power load according to claim 1, characterized in that: The optimization conditions include user experience comfort, carbon emissions, daily load, time-of-use electricity prices, and user equipment usage intentions.

9. A system for managing power consumption within a power load according to claim 8, characterized in that: The multi-objective optimization module adopts an optimization model of the load power management system based on time-of-use electricity prices and multi-objective optimization. The model classifies the loads and monitors the equipment, and achieves reasonable operation and an operating state within the load range through an improved particle swarm algorithm.

10. A method for implementing a power management system within a power load as described in any one of claims 1 to 9, characterized in that: The following steps are involved: (1) Data collection: The data collection module collects power load data in real time, including voltage, current, power factor, active power, reactive power and power output data, through sensors and smart meters installed at each power load point. The data collection module uploads the power load data to the data storage and processing module for processing through the data transmission module; (2) Data processing: The data storage and processing module cleans, supplements missing values, standardizes, normalizes, and fills missing values ​​on the collected power load data, thereby improving the quality and availability of the data; (3) Load optimization: The load prediction module transmits the load prediction data of the day to the multi-objective optimization control block. The multi-objective optimization control block connects the output state to the intelligent switch module and transmits the optimization result to the intelligent switch module, thereby controlling the load module to maintain the normal operation of different loads, thereby ensuring that the power load is within the allowable range.

Images