Household solar power generation and energy supply scheduling control method

Abstract

The invention discloses a household solar power generation and energy supply scheduling control method, and relates to the technical field of energy scheduling, and the method comprises the steps: obtaining a household historical power consumption data set; performing time domain load analysis on the historical electricity consumption data to obtain household historical electricity consumption load characteristic parameter data; performing load characteristic decomposition according to the household historical electricity consumption load characteristic parameter data to obtain electricity demands of a plurality of electric devices; dividing the electricity demands of the plurality of electric devices according to the operation instructions of the electric devices, analyzing the energy consumption conversion efficiency of the electricity demands corresponding to the operation instructions of the electric devices, and marking the electricity demand energy efficiency indexes of the operation instructions of the electric devices; and on the basis that the electricity demand energy efficiency index of the operation instruction of the electric equipment points to the electricity task type, an energy supply scheduling control decision model is constructed, and a household power generation energy supply scheduling control decision is generated. The system has the advantages that the household energy utilization efficiency is improved, and the electric charge expenditure is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of energy scheduling, and specifically relates to a household solar power generation energy supply scheduling control method. Background Art

[0002] Household solar power generation energy supply refers to a clean energy utilization method that uses solar panels to convert solar energy into electrical energy, and stores and distributes the electrical energy through a control system to provide power supply for households.

[0003] Since the existing household solar power generation energy supply strategy is to use the power immediately as it is generated, but the electricity demand of household electrical appliances varies greatly when performing different tasks. The electricity demand for the initial tasks of electrical appliances is large but the energy consumption efficiency is low. Therefore, by learning and identifying such tasks, giving priority to using the electricity generated by solar power, and switching to grid alternating current after the electrical appliances enter the stable stage, the household energy utilization efficiency can be improved and the electricity bill expenditure can be effectively reduced. Summary of the Invention

[0004] To solve the above technical problems, a household solar power generation energy supply scheduling control method is provided. This technical solution solves the problem that the existing household solar power generation energy supply strategy is to use the power immediately as it is generated, but the electricity demand of household electrical appliances varies greatly when performing different tasks, and the electricity demand for the initial tasks of electrical appliances is large but the energy consumption efficiency is low.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A household solar power generation energy supply scheduling control method, comprising:

[0007] Obtain household historical electricity consumption data and form a household historical electricity consumption dataset;

[0008] Based on the household historical electricity consumption dataset, perform time-domain load analysis on the historical electricity consumption data to obtain household historical electricity consumption load characteristic parameter data;

[0009] Decompose the load characteristics according to the household historical electricity consumption load characteristic parameter data to obtain the electricity demands of several electrical appliances;

[0010] Divide the electricity demands of several electrical appliances according to the operation instructions of the electrical appliances, analyze the energy consumption conversion efficiency of the electricity demands corresponding to the operation instructions of the electrical appliances, and mark the energy efficiency indicators of the electricity demands of the operation instructions of the electrical appliances;

[0011] Based on the energy efficiency indicators of the electricity demands of the operation instructions of the electrical appliances pointing to the electricity task types, construct an energy supply scheduling control decision model and generate a household power generation energy supply scheduling control decision.

[0012] Preferably, based on the household historical electricity consumption dataset, perform time-domain load analysis on the historical electricity consumption data to obtain the household historical electricity consumption load characteristic parameter data, specifically including:

[0013] Clean the data, process missing values and outliers based on the household historical electricity consumption dataset, and construct a household historical electricity consumption time series diagram;

[0014] Use the autocorrelation function to analyze the electricity consumption change trend at each time node in the household historical electricity consumption time series diagram to obtain the household electricity periodic characteristics;

[0015] Calculate the change trend of the electricity consumption time series index at each time node in the household historical electricity consumption time series diagram to obtain the household electricity time series characteristics; the electricity consumption time series index includes: average electricity consumption, variance of electricity consumption, standard deviation of electricity consumption;

[0016] Based on the time index of the household historical electricity consumption dataset, use the clustering algorithm, take the household electricity periodic characteristics as the clustering clusters, divide the household electricity time series characteristics, and establish a household historical electricity consumption time domain distribution diagram;

[0017] Among them, the clustering algorithm is specifically:

[0018]

[0019] In the formula, C(x i ) is the clustering function that inputs the household electricity time series characteristic x i and returns the index of the cluster to which the data point i belongs. dist(x i , μ j ) is the distance function that calculates the distance between the input household electricity time series characteristic x i and the centroid μ of the clustering cluster j . j is the index of the clustering cluster, and k is the total number of clustering clusters;

[0020] Based on the household historical electricity consumption time domain distribution diagram, use the Fourier algorithm to convert the household historical electricity time domain characteristic distribution into the frequency domain to obtain the household historical electricity consumption frequency domain distribution diagram;

[0021] According to the household historical electricity consumption frequency domain distribution diagram, extract the electricity consumption load characteristic parameters at each time node to obtain the household historical electricity consumption load characteristic parameter data; the electricity consumption load characteristic parameters include: electricity consumption load frequency components, electricity consumption load frequency distribution;

[0022] Among them, the Fourier algorithm is specifically:

[0023]

[0024] where y i is the complex amplitude of the i-th data point in the frequency component of the household historical electricity consumption time-domain distribution diagram, and y n is a complex number sequence of length n of the household historical electricity consumption time-domain distribution diagram, is the complex exponential function.

[0025] Preferably, load characteristic decomposition is performed according to the household historical electricity consumption load characteristic parameter data, and the electricity consumption demands of several electrical devices are obtained, specifically including:

[0026] Obtain big data on the electricity consumption characteristics during the operation of several household electrical devices;

[0027] Based on the power consumption identification random forest of household electrical devices, calculate the maximum information gain of the branch division threshold of the internal nodes of the power consumption identification decision tree of each household electrical device corresponding to the big data on the electricity consumption characteristics during the operation of several household electrical devices, and use it as the branch electricity consumption division characteristic of the internal nodes to form the power consumption identification random forest of household electrical devices;

[0028] Based on the power consumption identification random forest of household electrical devices, use the household historical electricity consumption load characteristic parameter data as the root node input and the electricity consumption demand of household electrical devices as the output;

[0029] Among them, the power consumption identification random forest of the household electrical devices is specifically:

[0030]

[0031] where U z is the electricity consumption demand of the z-th household electrical device, I() is the indicator function, and E v is the v-th power consumption identification decision tree of household electrical devices, and y ′ i′ is the household historical j ′ electricity consumption load characteristic parameter data, z is the category label of the electrical device, and m is the total number of power consumption identification decision trees of household electrical devices.

[0032] Preferably, based on the electricity consumption demand energy efficiency index of the operation instruction of the electrical device pointing to the electricity consumption task type, a power supply scheduling control decision model is constructed, and the household power generation supply scheduling control decision is generated, specifically including:

[0033] Obtain the actual power generation amount of household solar power generation;

[0034] Obtain the real-time electricity consumption task type of household electrical devices, and establish the electricity consumption demand energy efficiency index matrix A of the operation instructions of real-time electrical devices;

[0035]

[0036] Among them, a i′k′ is the electricity demand energy efficiency index of the k ′ th operation instruction of the i ′ th real-time electricity-consuming device, n ′ is the total number of real-time electricity-consuming devices, and k ′ is the total number of operation instructions;

[0037] Normalize the electricity demand energy efficiency index of the operation instructions for the electricity-consuming devices, and assign the operation instruction priority weight pointing to the electricity task type;

[0038] Based on each element in the electricity demand energy efficiency index matrix of the operation instructions of the real-time electricity-consuming devices and the priority weight of the operation instruction pointing to the electricity task type, calculate the allocated power of solar power generation for household electricity-consuming devices;

[0039] Construct a BP neural network and construct an energy supply scheduling control decision model;

[0040] Based on the actual generated power of household solar power generation, use the allocated power of solar power generation for real-time household electricity-consuming devices as a limiting condition;

[0041] Based on the energy supply scheduling control decision model, with the optimal allocation of the generated power of the optimal household solar power generation at each moment as the allocation target according to the limiting condition, use the actual generated power of household solar power generation and the real-time electricity task type of household electricity-consuming devices as inputs to generate an energy supply scheduling control decision for household solar power generation.

[0042] Among them, the specific calculation of the allocated power of solar power generation for household electricity-consuming devices is as follows:

[0043] P = ∑a i′k′ ×W k′

[0044] In the formula, P is the allocated power of solar power generation for household electricity-consuming devices, and W k′ is the k ′ th operation instruction priority weight pointing to the electricity task type.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] The present invention proposes a household solar power generation energy supply scheduling control scheme. By analyzing and learning historical household electricity consumption data, identifying the electricity demand of electricity-consuming devices and their energy consumption conversion efficiency, an energy supply scheduling control decision model is constructed to optimize the scheduling and use of household solar power generation and external alternating current. Its beneficial effects are that it can improve the household energy utilization efficiency and effectively reduce the electricity bill expenditure. Description of the Drawings

[0047] Figure 1 It is a flowchart of a household solar power generation energy supply scheduling control method;

[0048] Figure 2 It is a flowchart of a method for obtaining characteristic parameter data of household historical electricity consumption load;

[0049] Figure 3 It is a flowchart of a method for obtaining the electricity demand of several electrical devices;

[0050] Figure 4 It is a flowchart of a household power generation energy supply scheduling control decision method. Detailed Implementation Manner

[0051] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious variations.

[0052] Refer to Figure 1 As shown, a household solar power generation energy supply scheduling control method includes:

[0053] Obtain household historical electricity data and form a household historical electricity consumption data set;

[0054] Based on the household historical electricity consumption data set, perform time-domain load analysis on the historical electricity consumption data to obtain characteristic parameter data of the household historical electricity consumption load;

[0055] According to the characteristic parameter data of the household historical electricity consumption load, perform load characteristic decomposition to obtain the electricity demands of several electrical devices;

[0056] Divide the electricity demands of several electrical devices according to the operation instructions of the electrical devices, analyze the energy consumption conversion efficiency of the electricity demands corresponding to the operation instructions of the electrical devices, and mark the energy efficiency indicators of the electricity demands of the operation instructions of the electrical devices;

[0057] Based on the energy efficiency indicators of the electricity demands of the operation instructions of the electrical devices pointing to the electricity task types, construct an energy supply scheduling control decision model and generate a household power generation energy supply scheduling control decision.

[0058] It is understandable that existing household electrical appliances have a relatively high power consumption during startup and operation, while the power consumption is relatively low when the electrical appliances enter the stable operation period, and the energy consumption conversion efficiency also reaches the maximum. Therefore, after dividing the execution tasks according to the power consumption requirements corresponding to the execution instructions of the electrical appliances, the execution tasks with relatively high power consumption and low conversion efficiency are preferentially powered by solar energy, and converted to external AC energy when the execution tasks enter the stable operation period, so as to improve the efficiency of household energy use and reduce household electricity expenses.

[0059] The above solution analyzes and learns the historical household electricity consumption data, identifies the power consumption requirements and energy consumption conversion efficiency of electrical appliances, and constructs an energy supply scheduling control decision model to optimize the scheduling and use of household solar power generation and external alternating current. Its beneficial effect is that it can improve the efficiency of household energy utilization and effectively reduce electricity expenses.

[0060] Refer to Figure 2 As shown, based on the household historical electricity consumption dataset, a time-domain load analysis is performed on the historical electricity consumption data to obtain the household historical electricity consumption load characteristic parameter data, which specifically includes:

[0061] Clean the data, process missing values and outliers based on the household historical electricity consumption dataset, and construct a household historical electricity consumption time series diagram;

[0062] Use the autocorrelation function to analyze the change trend of electricity consumption at each time node in the household historical electricity consumption time series diagram to obtain the periodic characteristics of household electricity consumption;

[0063] Calculate the change trend of the electricity consumption time series index at each time node in the household historical electricity consumption time series diagram to obtain the time series characteristics of household electricity consumption; the electricity consumption time series index includes: average electricity consumption, variance of electricity consumption, standard deviation of electricity consumption;

[0064] Based on the time index of the household historical electricity consumption dataset, use the clustering algorithm, take the periodic characteristics of household electricity consumption as the clustering clusters, and divide the time series characteristics of household electricity consumption to establish a household historical electricity consumption time domain distribution map;

[0065] Among them, the specific clustering algorithm is:

[0066]

[0067] In the formula, C(x i ) is the clustering function that inputs the time series characteristics of household electricity consumption x i and returns the index of the cluster to which the data point i belongs. dist(x i ,μ j ) is the distance function, which calculates the input time series characteristics of household electricity consumption x iThe distance from the centroid μ of the cluster, where j is the index of the cluster and k is the total number of clusters; j and k is the total number of clusters;

[0068] Based on the household historical electricity consumption time-domain distribution map, using the Fourier algorithm, the household historical electricity consumption time-domain feature distribution is converted into the frequency domain to obtain the household historical electricity consumption frequency-domain distribution map;

[0069] According to the household historical electricity consumption frequency-domain distribution map, the electricity consumption load characteristic parameters at each time node are extracted to obtain the household historical electricity consumption load characteristic parameter data; the electricity consumption load characteristic parameters include: the frequency components of the electricity consumption load and the frequency distribution of the electricity consumption load;

[0070] Among them, the Fourier algorithm is specifically:

[0071]

[0072] In the formula, y i is the complex amplitude of the i-th data point in the household historical electricity consumption time-domain distribution map on the frequency component, and y n is a complex sequence of length n of the household historical electricity consumption time-domain distribution map, is the complex exponential function.

[0073] By cleaning the household historical electricity consumption data, dealing with missing values and outliers, constructing a time series graph to show the electricity consumption situation; using the autocorrelation function to analyze the periodic characteristics, calculating time series indicators to obtain time series characteristics; using the clustering algorithm to divide the household electricity consumption characteristics and establish a time-domain distribution map; finally, converting to the frequency domain through the Fourier algorithm and extracting the electricity consumption load characteristic parameters. Accurately quantify the periodic, time series and frequency characteristics of household electricity consumption.

[0074] Referring to Figure 3 as shown, according to the household historical electricity consumption load characteristic parameter data, load characteristic decomposition is carried out to obtain the electricity consumption demands of several electrical devices, specifically including:

[0075] Obtain the big data of the electricity consumption characteristics during the operation of several household electrical devices;

[0076] Based on the power consumption of household electrical devices to identify the random forest, calculate the maximum information gain of the branch division threshold of the internal nodes of the decision tree for the power consumption identification of each household electrical device corresponding to the big data of the electricity consumption characteristics during the operation of several household electrical devices, and use it as the branch electricity consumption division characteristic of the internal nodes to form the random forest for the power consumption identification of household electrical devices;

[0077] A random forest for identifying the power consumption of household electrical appliances takes the characteristic parameter data of historical household power consumption load as the input of the root node and the power consumption demand of household electrical appliances as the output;

[0078] Among them, the random forest for identifying the power consumption of household electrical appliances is specifically as follows:

[0079]

[0080] In the formula, U z is the power consumption demand of the z-th household electrical appliance, I() is the indicator function, and E v is the v-th decision tree for identifying the power consumption of household electrical appliances, and y ′ i′ is the characteristic parameter data of the historical household power consumption load at the j-th ′ time, z is the category label of the electrical appliance, and m is the total number of decision trees for identifying the power consumption of household electrical appliances.

[0081] In this solution, by collecting the big data of the power consumption characteristics of household electrical appliances and using the random forest algorithm to construct a power consumption identification model for household electrical appliances, this model determines the power consumption division characteristics by calculating the maximum information gain of the internal node branch division threshold, and then inputs the characteristic parameter data of the historical household power consumption load into the model to output the power consumption demand of each electrical appliance. The beneficial effect is that it can accurately decompose the household power load, identify the power consumption demand of each device, and provide core parameters for the subsequent steps.

[0082] Refer to Figure 4 As shown, based on the power consumption energy efficiency index of the operation instruction of the electrical appliance pointing to the power consumption task type, constructing an energy supply scheduling control decision model, and generating the household power generation energy supply scheduling control decision specifically includes:

[0083] Obtain the actual power generation amount of household power generation solar energy;

[0084] Obtain the real-time power consumption task type of household electrical appliances, and establish a power consumption energy efficiency index matrix A of the operation instructions of real-time electrical appliances;

[0085]

[0086] Among them, a i′k′ is the power consumption energy efficiency index of the k-th operation instruction of the i-th ′ real-time electrical appliance, n ′ is the total number of real-time electrical appliances, and k ′ is the total number of operation instructions; ′

[0087] ​Normalize the energy efficiency index of the power consumption demand for the operation instructions of the electrical equipment, and assign the operation instruction priority weight pointing to the type of power consumption task;

[0088] Based on each element in the energy efficiency index matrix of the operation instructions of the real-time electrical equipment and the priority weight of the operation instruction pointing to the type of power consumption task, calculate the allocated power generation of solar power for household electrical equipment;

[0089] Construct a BP neural network and construct an energy supply scheduling control decision model;

[0090] Based on the actual power generation of household solar power generation, use the allocated power generation of solar power for real-time household electrical equipment as a constraint condition;

[0091] Based on the energy supply scheduling control decision model, with the optimal power generation allocation of the optimal household solar power generation at each moment as the allocation target according to the constraint condition, use the actual power generation of household solar power generation and the real-time power consumption task type of household electrical equipment as inputs to generate an energy supply scheduling control decision for household solar power generation;

[0092] Among them, the specific calculation of the allocated power generation of solar power for household electrical equipment is as follows:

[0093] P = ∑a i′k′ ×W k′

[0094] In the formula, P is the allocated power generation of solar power for household electrical equipment, and W k′ is the k ′ th operation instruction priority weight pointing to the type of power consumption task.

[0095] This solution constructs an energy supply scheduling control decision model by constructing a BP neural network. According to the real-time power consumption task type and energy efficiency index of the electrical equipment, as well as the actual power generation of household solar power generation, dynamically allocate the power generation of solar power, maximize the satisfaction of household power consumption demand, and improve energy efficiency utilization and economic benefits.

[0096] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for dispatching and controlling household solar power generation, characterized in that: include: Obtain historical household electricity consumption data and build a historical household electricity consumption data set; Based on the household historical electricity consumption data set, time domain load analysis is performed on the historical electricity consumption data to obtain the household historical electricity consumption load characteristic parameter data; Decompose the load characteristics according to the historical household electricity consumption load characteristic parameter data to obtain the electricity demand of several electrical equipment; Divide the power demand of several power-consuming devices according to the operation instructions of the power-consuming devices, analyze the energy consumption conversion efficiency between the operation instructions of the power-consuming devices and the power demand, and mark the energy efficiency index of the power demand of the operation instructions of the power-consuming devices; The power demand energy efficiency index based on the operating instructions of the power-consuming equipment points to the power consumption task type, builds an energy supply scheduling and control decision model, and generates household power generation energy supply scheduling and control decisions.

2. A household solar power generation energy supply scheduling control method according to claim 1, characterized in that: Based on the household historical electricity consumption data set, time domain load analysis is performed on the historical electricity consumption data, and the household historical electricity consumption load characteristic parameter data is obtained, including: Based on the household historical electricity consumption dataset, we cleaned the data, processed missing values ​​and outliers, and constructed a time series diagram of household historical electricity consumption; Using the autocorrelation function, the electricity consumption trend of each time node in the household historical electricity consumption time series diagram is analyzed to obtain the periodic characteristics of household electricity consumption; Calculate the change trend of the electricity consumption time series indicators at each time node in the household historical electricity consumption time series diagram to obtain the household electricity consumption time series characteristics; the electricity consumption time series indicators include: the mean of electricity consumption, the variance of electricity consumption, and the standard deviation of electricity consumption; Based on the time index of the household historical electricity consumption data set, a clustering algorithm is used to take the periodic characteristics of household electricity consumption as clusters, divide the household electricity consumption time series characteristics, and establish a time domain distribution map of household historical electricity consumption; The clustering algorithm is specifically: In the formula, C(x i ) is the clustering function input household electricity time series feature x i And returns the index of the cluster to which the data point i belongs, dist(x i ,μ j ) is the distance function, and the input household electricity time series feature x is calculated. i and the cluster centroid μ j The distance between them, j is the index of the cluster, k is the total number of clusters; Based on the time domain distribution diagram of household historical electricity consumption, the time domain characteristic distribution of household historical electricity consumption is converted into the frequency domain by using Fourier algorithm to obtain the frequency domain distribution diagram of household historical electricity consumption; According to the household historical electricity consumption frequency domain distribution diagram, the electricity consumption load characteristic parameters of each time node are extracted to obtain the household historical electricity consumption load characteristic parameter data; the electricity consumption load characteristic parameters include: electricity consumption load frequency component and electricity consumption load frequency distribution.

3. A household solar power generation energy supply scheduling control method according to claim 2, characterized in that: The Fourier algorithm is specifically: In the formula, y i is the complex amplitude of the frequency component of the ith data point in the time domain distribution diagram of household electricity consumption history, y n The time domain distribution graph of household historical electricity consumption is a complex sequence with a length of n. is the complex exponential function.

4. A household solar power generation energy supply scheduling control method according to claim 3, characterized in that: According to the historical household electricity consumption load characteristic parameter data, the load characteristics are decomposed to obtain the power demand of several electrical equipment, including: Obtaining big data on the power consumption characteristics of several types of household power equipment during operation; Based on the random forest for identifying power consumption of household electrical appliances, the maximum value of the information gain of the branch division threshold of the internal nodes of the power consumption identification decision tree of each type of household electrical appliance corresponding to the power consumption feature big data of several types of household electrical appliances during operation is calculated, and the maximum value is used as the branch power consumption division feature of the internal node to form a random forest for identifying power consumption of household electrical appliances; Based on the random forest for identifying power consumption of household electrical appliances, the characteristic parameter data of household historical power consumption load is used as the root node input, and the power demand of household electrical appliances is used as the output.

5. A household solar power generation energy supply scheduling control method according to claim 4, characterized in that: The random forest for identifying power consumption of household electrical appliances is specifically: Where U z is the power demand of the zth household electrical appliance, I() is the indicator function, E v is the power consumption identification decision tree of the vth household electrical appliance, y ′ i′ For home history ′ The characteristic parameter data of electricity consumption load, z is the category label of the electrical equipment, and m is the total number of power consumption identification decision trees of household electrical equipment.

6. A household solar power generation energy supply scheduling control method according to claim 5, characterized in that: Based on the power demand energy efficiency index of the operating instructions of the power-consuming equipment, the power supply dispatching and control decision model is constructed to generate the household power generation power supply dispatching and control decision, which specifically includes: Get the actual amount of electricity generated by solar power for household power generation; Obtain the real-time power consumption task type of household electrical equipment and establish the power demand energy efficiency index matrix A of the real-time power consumption equipment operation instructions; Among them, a i′k′ For the i ′ The kth real-time power consumption device ′ The power demand energy efficiency index of the operation instruction, n ′ is the total number of real-time power-consuming devices, k ′ is the total number of running instructions; Normalize the power demand energy efficiency index of the operation instructions of the power-consuming equipment, and assign priority weights to the operation instructions pointing to the power-consuming task type; Based on each element in the power demand energy efficiency index matrix of the real-time power-consuming equipment operation instructions and the priority weight of the power-consuming task type, the solar power generation distribution power of the household power-consuming equipment is calculated; Construct BP neural network and energy supply dispatch control decision model; Based on the actual amount of solar power generated by household electricity generation, the amount of solar power generated by household electrical equipment in real time is used as a restriction condition; Based on the energy supply scheduling and control decision model, the optimal allocation of household solar power generation at each moment is taken as the allocation target according to the constraints, and the actual power generation of household solar power generation and the real-time power consumption task type of household electrical equipment are taken as input to generate household power supply scheduling and control decisions.

7. A household solar power generation energy supply scheduling control method according to claim 6, characterized in that: The calculation of the solar power generation distribution amount of household electrical equipment is specifically as follows: P=∑a i′k′ ×W k′ Where P is the solar power generation distribution of household electrical equipment, W k′ is the kth type of electricity consumption task ′ The priority weight of each running instruction.

Images