A load power consumption management and control method and system based on time-of-use electricity price

Abstract

The present invention relates to the field of power management, and specifically to a load power consumption management and control method and system based on time-of-use electricity prices. The method comprises: obtaining historical power consumption data of each user and constructing a fitness function for a load scheduling plan; setting an initial population; calculating the coefficient of variation of each user in each time period; for each individual, weighting a preset initial mutation rate by the coefficient of variation of each user in each time period to obtain a weighted mutation rate of each user in each time period; applying the weighted mutation rate to iteratively update the initial population, outputting the individual with the highest fitness in the current population as the optimal load scheduling plan, and optimizing user load scheduling based on the optimal load scheduling plan. By dynamically adjusting the mutation rate, the present invention can more quickly identify excellent individuals during the iterative process, thereby reducing redundant calculations, improving the efficiency of the algorithm, and enabling the algorithm to promptly provide the optimal scheduling plan.

Description

Technical Field

[0001] The present invention relates to the field of power management, and more particularly to a load power consumption management and control method and system based on time-of-use electricity pricing. Background Art

[0002] Time-of-use electricity pricing is a mechanism that divides the 24-hour day into multiple periods based on power system load fluctuations, with different electricity prices set for each period. This pricing signal guides users to adjust their electricity usage, encouraging them to use more electricity during periods of low load and less during peak hours. This achieves the effect of peak load shifting, ensuring the safe and stable operation of the power system and improving overall system efficiency.

[0003] Prior art, such as the patent application document with publication number CN116012078A, discloses an electricity price optimization method based on a genetic algorithm, which includes: obtaining M original populations, each of which corresponds to a preset time period on the previous day, and the original population includes multiple individuals, and the multiple individuals are the initial electricity prices at different times in the preset time period on the previous day; for each of the original populations, using a genetic algorithm to iteratively update the original population, and obtaining a target population when the iteration termination condition is met, the target population includes multiple target sub-individuals, and the target sub-individuals are updated electricity prices after updating the original electricity prices corresponding to some individuals in the original population; based on each of the target sub-individuals in each of the target populations, determining the optimized electricity price for each preset time period on the day.

[0004] However, the genetic algorithm involved in the above electricity price optimization method has a mutation rate that is too large or too small, resulting in redundant calculations or low convergence efficiency in the optimization process of user load scheduling schemes with different controllable loads. Summary of the Invention

[0005] In order to solve the technical problem that the algorithm cannot provide the optimal scheduling solution in time due to redundant calculation or slow convergence caused by improper mutation rate setting, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a load power consumption management and control method based on time-of-use electricity prices includes:

[0007] Obtain historical electricity consumption data of each user and construct a fitness function for the load scheduling plan;

[0008] Optimizing the fitness function to obtain an optimal load scheduling solution;

[0009] Optimize user load scheduling based on the optimal load scheduling plan;

[0010] The optimization process includes:

[0011] Setting an initial population, where the initial population consists of multiple individuals, each of which represents a load scheduling scheme;

[0012] Calculate the coefficient of variation for each user in each time period; for each individual, weight the preset initial variation rate by the coefficient of variation for each user in each time period to obtain the weighted variation rate for each user in each time period; the coefficient of variation is positively correlated with the user's controllability over electricity consumption and negatively correlated with the stability of electricity consumption;

[0013] The weighted mutation rate is applied to iteratively update the initial population, and the individual with the highest fitness in the current population is output as the optimal load scheduling solution.

[0014] The present invention first obtains the historical electricity consumption data of each user, so as to more accurately understand the user's electricity consumption needs and habits, and provide a solid data foundation for constructing the fitness function of the load scheduling scheme. Further, by optimizing the load scheduling scheme, the supply and demand relationship of the power system can be better balanced, and the overload or overload of the power system can be avoided, thereby enhancing the stability of the power system. In the optimization process, the user's controllability and stability of electricity consumption are taken into account, and a weighted mutation rate is calculated. The weighted mutation rate allows the algorithm to dynamically adjust its parameters according to the user's actual needs and behavior. In the iterative process, excellent individuals can be identified more quickly and individuals with poor performance can be eliminated, thereby reducing redundant calculations, improving the efficiency of the algorithm, and enabling the algorithm to provide the optimal scheduling plan in a timely manner.

[0015] Preferably, the coefficient of variation satisfies the relationship:

[0016] Where, For users In the The coefficient of variation for each period, For users In the The degree of controllability of electricity consumption in each period, For users In the The stability of electricity consumption during each period.

[0017] By introducing the controllability and stability of electricity consumption, the calculation formula of the coefficient of variation can comprehensively consider the adjustability and stability of users' electricity consumption behavior, thereby more comprehensively evaluating the characteristics of users' electricity consumption behavior in specific time periods, and providing important reference for the optimization, management and decision-making of the power system.

[0018] Preferably, the controllability satisfies the relationship:

[0019] Where, For users In the The degree of controllability of electricity consumption in each period, For users In the The controllable factors of electricity consumption in each period, 、 These are all pre-set parameters. It is an exponential function with the natural constant e as its base.

[0020] The formula nonlinearly maps the controllable factors and controllability of electricity consumption in the form of an exponential function, which can more flexibly reflect changes in users' electricity consumption behavior; by calculating the controllability of different users in different time periods, the power system can more accurately identify which users are more likely to respond to regulation needs in which time periods.

[0021] Preferably, the stability satisfies the relationship:

[0022] Where, For users In the The stability of electricity consumption in each period, For the history Daily users In the The maximum power consumption in each period, For the history Daily users In the The minimum power consumption in each period, Indicates normalization processing.

[0023] By calculating the difference between the maximum and minimum power consumption in the same period of historical data and normalizing it, we can get a specific value to represent the user's power consumption stability during that period. The smaller this value, the more stable the user's power consumption during that period.

[0024] Preferably, the iterative update includes:

[0025] According to the value of the fitness function, some individuals are selected from the current population to enter the next generation;

[0026] Perform crossover operations on the selected individuals to simulate the hybridization process in the biological world and produce new offspring individuals;

[0027] Performing a mutation operation on the offspring individuals according to the weighted mutation rate to generate new offspring individuals;

[0028] The selection, crossover and mutation operations are repeated to form a new generation of population; when the preset maximum number of iterations is reached, the iteration is terminated.

[0029] Preferably, the process of obtaining the controllable power consumption factor includes:

[0030] Calculate the deviation between the electricity consumption of any user in each time period and the historical average electricity consumption;

[0031] Calculate the deviation between the electricity price of any user in each period and the historical average electricity price;

[0032] Multiply the normalized value of the deviation between the user's electricity consumption in each time period and the historical average electricity consumption and the deviation between the electricity price in each time period and the historical average electricity price, and then divide it by the total number of days of historical data to obtain the user's electricity consumption controllable factor in the corresponding time period.

[0033] By multiplying the normalized values of electricity consumption deviation and electricity price deviation and dividing them by the total number of days of historical data, we can obtain an indicator that comprehensively considers changes in electricity consumption and electricity prices. This indicator reflects users' sensitivity to electricity price changes during a specific period of time and their ability to adjust their electricity consumption behavior.

[0034] Preferably, the process of obtaining the stability includes:

[0035] Calculate the maximum and minimum power consumption of the user in the current time period on any day in history, and normalize the product of the difference between the maximum and minimum power consumption and the absolute value of the autocorrelation coefficient of the user's power consumption in the current time period as the stability of the user's power consumption in the current time period.

[0036] In the second aspect, a load power consumption management and control system based on time-of-use electricity prices includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, any one of the load power consumption management and control methods based on time-of-use electricity prices is implemented.

[0037] The present invention has the following effects:

[0038] The present invention uses an intelligent optimization algorithm, combined with users' historical electricity consumption data and time-of-use electricity price information, to achieve efficient management of load scheduling. It not only improves convergence efficiency and reduces redundant calculations, but also provides users with practical electricity usage suggestions, helping to build a greener, more efficient and reliable power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0040] Figure 1 This is a method flow chart of steps S1 to S3 in a load power consumption management and control method based on time-of-use electricity prices in an embodiment of the present invention.

[0041] Figure 2 It is a schematic diagram of steps S20 to S22 in a load power consumption management and control method based on time-of-use electricity prices in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0043] The application scenario of the present invention is: using the improved genetic algorithm to find the optimal load scheduling solution for users under time-of-use electricity prices.

[0044] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0045] Reference Figure 1 A load power management and control method based on time-of-use electricity prices includes steps S1 to S3, which are specifically as follows:

[0046] S1: Obtain historical electricity consumption data of each user and construct a fitness function for the load scheduling scheme.

[0047] The user management system collects historical electricity usage data for each user, including daily, monthly, or annual electricity usage, peak and off-peak periods, and other information. In an embodiment of the present invention, the historical electricity usage data for each user is obtained from the user management system, and the data includes electricity usage for each period in the past 30 days.

[0048] Clean and organize historical electricity consumption data, remove outliers, fill missing values and perform normalization.

[0049] In addition, a day is divided into 24 time periods, and the average electricity price of each time period is obtained as the electricity price of that time period.

[0050] Because electricity prices vary across time periods, some may be higher and others lower. To minimize user costs, users should be encouraged to use electricity during times when prices are lower and reduce their usage during times when prices are higher. At the same time, load balancing within the power system must be ensured. Load balancing means that electricity consumption across users should be balanced across time periods, avoiding excessive power consumption in some periods and insufficient power consumption in others. This ensures stable operation of the power system, avoids power shortages or oversupply during certain periods, and improves the efficiency of power resource utilization.

[0051] In summary, when developing a load dispatch plan, it's necessary to simultaneously consider reducing user electricity costs and ensuring load balancing across the power system. These two objectives can sometimes conflict with each other, so intelligent optimization algorithms are needed to find an optimal dispatch plan that strikes the right balance between these two goals.

[0052] Furthermore, in order to find the optimal load scheduling solution, it is necessary to construct a fitness function that comprehensively considers the total electricity bill of the user and the degree of load balancing.

[0053] Specifically, the total number of users is obtained from the user management system, the total electricity charges of each user in all time periods are calculated, and the variance of the electricity consumption of all users in each time period, that is, the degree of load balancing, is calculated.

[0054] The fitness function of the load scheduling scheme is constructed based on the above-mentioned total electricity cost and load balancing degree, which satisfies the relationship:

[0055]

[0056] Where, is the fitness of the load dispatching scheme, is the total number of users, is the total number of divided time periods (24 in this embodiment of the present invention), For the The electricity price for each period, For users In the The electricity consumption during each period, represents the normalization function, Represents the exponential function.

[0057] in, Represents a user The total electricity cost for all periods of time, It represents the variance of electricity consumption of all users in each time period.

[0058] S2: Optimize the fitness function to obtain an optimal load scheduling solution.

[0059] As an optimization algorithm that simulates the natural evolution process, genetic algorithm has significant advantages, such as strong global search capability, strong adaptability, parallel processing capability, high robustness, etc. It can handle various complex, nonlinear, and multidimensional optimization problems.

[0060] In genetic algorithms, the mutation rate is a key parameter controlling population diversity and search power. Users with flexible electricity usage schedules have more controllable loads, meaning they have more degrees of freedom in optimizing the solution space. However, their electricity usage habits may make their consumption more stable during certain periods, making adjustments difficult. During these periods, the degrees of freedom for optimizing are lower. To avoid redundant computation for users with less controllable loads when the mutation rate is too high (because their loads are difficult to change regardless of adjustments), and slow optimization for users with more controllable loads when the mutation rate is too low (because they require more time to explore the larger solution space), we need to understand the controllability of each user's electricity consumption and the stability of their consumption over different time periods. By identifying the degrees of freedom for optimizing each user in the solution space during each time period, we can dynamically adjust the mutation rate: when a user has a high degree of freedom for optimizing during a certain period, we assign a higher mutation rate to accelerate the search; conversely, we lower the mutation rate to avoid inefficient searches.

[0061] Reference Figure 2 The process of optimizing the fitness function includes steps S20 to S22, which are as follows:

[0062] S20: Setting an initial population, where the initial population consists of a plurality of individuals, each of which represents a load scheduling scheme.

[0063] In a genetic algorithm, an initial population must first be generated. This population consists of multiple individuals, each representing a possible solution.

[0064] Specifically, the gene range is first determined. In the embodiment of the present invention, the minimum value is set to 0, the maximum value is set to twice the maximum power consumption in the historical power consumption data of all users, and the minimum unit of the solution space is two decimal places.

[0065] Then, within the above range, a population of 500 individuals is randomly generated, each of which represents a load scheduling plan, including the electricity consumption of each user in each time period.

[0066] S21: Calculate the coefficient of variation for each user in each time period; for each individual, weight the preset initial variation rate with the coefficient of variation for each user in each time period to obtain the weighted variation rate for each user in each time period; the coefficient of variation is positively correlated with the user's controllability over electricity consumption and negatively correlated with the stability of electricity consumption.

[0067] By analyzing users' electricity usage behaviors and habits, we determine the controllability and stability of their electricity usage over different time periods. We then use this information to adjust each user's mutation rate to more effectively find the optimal electricity allocation solution in the solution space. This improves search efficiency, avoids unnecessary calculations, and ultimately achieves more efficient power management.

[0068] In an embodiment of the present invention, first, for each user in each time period, the deviation of his / her electricity consumption from the historical average electricity consumption (i.e., the variation of electricity consumption) and the deviation of the electricity price corresponding to each time period from the historical average electricity price (i.e., the variation of electricity price) are calculated. Then, the variation of electricity consumption is multiplied by the normalized value of the variation of electricity price, and all historical electricity consumption data are summed up. Finally, the electricity controllable factor of the user in the corresponding time period is obtained by dividing it by the total number of days of historical data. The larger the electricity controllable factor, the more sensitive the user is to the change in electricity price during the time period.

[0069] Then the above-mentioned controllable factors of electricity consumption satisfy the relationship:

[0070]

[0071] Where, For users In the The controllable factors of electricity consumption in each period, For the history Daily users In the The electricity consumption during each period, is the total number of historical days collected, For the history Tiandi The electricity price for each period, Indicates normalization processing.

[0072] When the historical electricity consumption data, users In the When the electricity consumption of a period increases more than the historical average electricity consumption of the same period, and the electricity price decreases more than the historical electricity price of the same period, the user In the For example, if the electricity consumption is high and the electricity price is low, or if the electricity consumption is low and the electricity price is high, this factor will be close to 1 or 0.

[0073] The above-mentioned controllable power consumption factor is further used to quantify the controllability of the user's power consumption in a specific period of time, that is, the relationship is satisfied:

[0074]

[0075] Where, For users In the The degree of controllability of electricity consumption in each period, For users In the The controllable factors of electricity consumption in each period, 、 These are all pre-set parameters. is an exponential function with the natural constant e as the base. The empirical value is set to 30 to adjust the shape of the formula to ensure that when the power controllability factor approaches 0 or 1, the power controllability gradually approaches 1; When the power consumption controllability factor is at an intermediate value (such as 0.5), the power consumption controllability reaches the minimum value of 0. This reflects that when the power consumption behavior is close to the average behavior, the user's power consumption controllability is the lowest.

[0076] Among them, when near hour, The value of is small, and the result of the exponential function is close to 1, so Close to 0, it means that the controllability of electricity consumption is low; on the contrary, when keep away (i.e. close to 0 or 1), As the value of is large, the result of the exponential function decreases, so A value close to 1 indicates a high degree of controllability of electricity consumption.

[0077] It should be noted that for users with a small fluctuation range in electricity consumption, since their electricity consumption is relatively stable, it is more difficult to adjust their electricity consumption (for example, to save electricity bills or respond to grid demand). This is because these users have formed relatively fixed electricity consumption habits, which are relatively difficult to change.

[0078] Therefore, the stability of the user's electricity consumption during a specific period is evaluated by calculating the difference between the user's maximum and minimum electricity consumption during the period and performing linear normalization on it.

[0079] Then the above stability satisfies the relationship:

[0080]

[0081] Where, For users In the The stability of electricity consumption in each period, For the history Daily users In the The maximum power consumption in each period, For the history Daily users In the The minimum power consumption in each period, Indicates normalization processing.

[0082] By calculating the stability of a user's electricity consumption during a specific period, we can understand the fluctuations in their electricity consumption behavior. The more stable a user is, the smaller their electricity consumption fluctuations are, and the easier it is to adjust their electricity consumption.

[0083] Then, according to the above and Calculate the user In the The coefficient of variation of each period satisfies the following relationship:

[0084]

[0085] Where, For users In the The coefficient of variation for each period, For users In the The degree of controllability of electricity consumption in each period, For users In the The stability of electricity consumption during each period.

[0086] According to the above user In the The calculation method of the coefficient of variation of each period is similar to the calculation of the user The coefficient of variation in other time periods and the coefficient of variation of other users in each time period.

[0087] Finally, the initial mutation rate in the genetic algorithm is preset to 0.3 based on the empirical value, and the preset initial mutation rate is weighted by the coefficient of variation of each user in each time period to obtain the weighted mutation rate of each user in each time period.

[0088] For example, still based on the user For example, the user In the The weighted mutation rate of each period satisfies the following relationship:

[0089]

[0090] Where, For users In the The weighted mutation rate of each period, For users In the The coefficient of variation for each period, is the initial mutation rate, is the total number of time periods, is the total number of users.

[0091] According to the user In the The weighted mutation rate of each time period can be calculated by the same method to get the weighted mutation rate of all users in all time periods.

[0092] S22: Apply the weighted mutation rate to iteratively update the initial population, and output the individual with the highest fitness in the current population as the optimal load scheduling solution.

[0093] Specifically, in the embodiment of the present invention, the weighted mutation rate calculated in step S21 is introduced into the genetic algorithm. Then, the general steps of searching for the optimal load scheduling solution according to the genetic algorithm are as follows:

[0094] Based on the value of the fitness function, some individuals are selected from the current population to enter the next generation. Common selection methods include roulette wheel selection and tournament selection.

[0095] Perform crossover operations on the selected individuals to simulate the hybridization process in the biological world and produce new offspring individuals;

[0096] Performing a mutation operation on the offspring individuals according to the weighted mutation rate to generate new offspring individuals;

[0097] Repeat the selection, crossover, and mutation operations to form a new generation of population. When the preset maximum number of iterations is reached (for example, 100 iterations), the iteration is terminated.

[0098] In the final population after the iteration stops, find the individual with the highest fitness, which is the optimal load scheduling plan. The plan includes the optimal power consumption arrangement of each user in each time period, as well as the corresponding recommended power consumption and other information.

[0099] S3: Optimize user load scheduling according to the optimal load scheduling plan.

[0100] According to the optimal load scheduling plan, the recommended power consumption information is directly delivered to users by sending text messages.

[0101] In another embodiment, the user In the The calculation formula of the coefficient of variation of each period is different. Considering the contribution of the controllability and stability of electricity consumption, the weight coefficient is introduced. In the The coefficient of variation of each period satisfies the relationship:

[0102]

[0103] Where, For users In the The coefficient of variation for each period, For users In the The degree of controllability of electricity consumption in each period, For users In the The stability of electricity consumption during each period.

[0104] 、 are weight coefficients, which can be determined based on experience, experimental data or expert opinions. For example, if we believe that the controllability of electricity consumption has a greater impact on the coefficient of variation, we can set a larger value; and vice versa.

[0105] In another embodiment, the user In the The calculation methods of the stability of electricity consumption in different time periods are different:

[0106]

[0107] Where, For users In the The stability of electricity consumption in each period, For the history Daily users In the The maximum power consumption in each period, For the history Daily users In the The minimum power consumption in each period, represents normalization processing, Represents a user In the The absolute value of the autocorrelation coefficient of electricity consumption in each period.

[0108] The system includes a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, the load power consumption management and control method based on time-of-use electricity price according to the first aspect of the present invention is implemented.

[0109] The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and therefore will not be described in detail here.

[0110] In the present invention, the aforementioned memory can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connected to the device.

[0111] In the description of this specification, "multiple" and "several" mean at least two, such as two, three or more, unless otherwise clearly defined.

[0112] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A load power management and control method based on time-of-use electricity price, characterized in that: include: Obtain historical electricity consumption data of each user and construct a fitness function for the load scheduling plan; Optimizing the fitness function to obtain an optimal load scheduling solution; Optimize user load scheduling according to the optimal load scheduling plan; The optimization process includes: Setting an initial population, where the initial population consists of multiple individuals, each of which represents a load scheduling scheme; Calculate the coefficient of variation for each user in each time period; for each individual, weight the preset initial variation rate by the coefficient of variation for each user in each time period to obtain the weighted variation rate for each user in each time period; the coefficient of variation is positively correlated with the user's controllability over electricity consumption and negatively correlated with the stability of electricity consumption; The weighted mutation rate is applied to iteratively update the initial population, and the individual with the highest fitness in the current population is output as the optimal load scheduling solution.

2. The load power management and control method based on time-of-use electricity price according to claim 1, characterized in that: The coefficient of variation satisfies the relationship: Where, For users In the The coefficient of variation for each period, For users In the The degree of controllability of electricity consumption in each period, For users In the The stability of electricity consumption during each period.

3. The load power management and control method based on time-of-use electricity price according to claim 2 is characterized in that: The controllability satisfies the relationship: Where, For users In the The degree of controllability of electricity consumption in each period, For users In the The controllable factors of electricity consumption in each period, 、 These are all pre-set parameters. It is an exponential function with the natural constant e as its base.

4. The load power management and control method based on time-of-use electricity price according to claim 3 is characterized in that: The iterative update includes: According to the value of the fitness function, some individuals are selected from the current population to enter the next generation; Perform crossover operations on the selected individuals to simulate the hybridization process in the biological world and produce new offspring individuals; Performing a mutation operation on the offspring individuals according to the weighted mutation rate to generate new offspring individuals; The selection, crossover and mutation operations are repeated to form a new generation of population; when the preset maximum number of iterations is reached, the iteration is terminated.

5. The load power management and control method based on time-of-use electricity price according to claim 4 is characterized in that: The process of obtaining the controllable power consumption factor includes: Calculate the deviation between the electricity consumption of any user in each time period and the historical average electricity consumption; Calculate the deviation between the electricity price of any user in each period and the historical average electricity price; Multiply the normalized value of the deviation between the user's electricity consumption in each time period and the historical average electricity consumption and the deviation between the electricity price in each time period and the historical average electricity price, and then divide it by the total number of days of historical data to obtain the user's electricity consumption controllable factor in the corresponding time period.

6. A load power management and control system based on time-of-use electricity prices, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the load power consumption management and control method based on time-of-use electricity price according to any one of claims 1 to 5 is implemented.

Images