Adjusting method of demand side response scheduling invitation considering maximum demand

Abstract

The invention relates to the technical field of participation of a virtual power plant in load resource demand response, and discloses a demand side response scheduling invitation adjustment method considering maximum demand, comprising the following steps: obtaining user load prediction influence data, and inputting the data into a user load prediction model to obtain a user electrical load prediction curve; collecting energy storage equipment information of the user and prepared demand response invitation information; calculating the maximum demand of the user according to the user electrical load prediction curve and the user electrical load actual curve, calculating an output curve of user energy storage equipment in a response time period, and further calculating a user response capability curve, thereby adjusting demand response invitation information, and performing response declaration by the adjusted demand response invitation; in the invitation response time period, the actual electrical load of the user is monitored in real time, the output condition of the energy storage equipment is regulated and controlled, and it is ensured that the maximum demand limit is not broken through; the technical problems that a virtual power plant is low in invitation response efficiency and poor in power generation side load fluctuation balancing capacity are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants participating in load resource demand response, and particularly relates to a regulation method for a demand-side response scheduling invitation considering the maximum demand. Background Art

[0002] With the transformation of the global energy structure, the installed capacity of high-proportion new energy is increasing day by day. For example, the large-scale access of renewable energy such as solar energy and wind energy. These new energies have the characteristics of intermittency and volatility. This instability leads to an increase in the volatility of the traditional power system, bringing huge challenges to the safe and stable operation of the power grid.

[0003] In order to increase the flexible regulation ability of the new power system and give full play to the frequency modulation and peak shaving functions of the user-side resources, the concept of virtual power plants has emerged. It can integrate and coordinate the distributed energy resources on the user side, enabling them to have a regulation ability similar to that of traditional power plants, thereby effectively balancing the load fluctuations on the power generation side and improving the stability and reliability of the power system.

[0004] Demand-side response refers to the behavior of power users to actively change their conventional electricity consumption patterns according to market price signals or incentive mechanisms. Essentially, virtual power plants achieve the goal of balancing the fluctuations on the power generation side through demand-side response, and at the same time, users on the power consumption side can also obtain economic benefits from it. On the one hand, users can change their electricity consumption behavior by participating in demand-side response, using electricity when the electricity price is cheap or under the incentive of response subsidies, reducing the electricity cost or obtaining additional income; on the other hand, the power grid can improve the operation efficiency and stability of the power system through the regulation of virtual power plants, reduce the demand for reserve capacity, and lower the operating cost.

[0005] However, the current response invitation methods of virtual power plants often do not fully consider the impact of response invitations on the electricity demand of the executing users. Since the establishment of virtual power plants is mainly to solve the problems mainly based on the power grid, in the actual implementation process, excessive emphasis is placed on the efficiency of response invitations to give full play to the full capacity of user loads.

[0006] In this mode, users may significantly increase or decrease their electricity loads when responding to invitations, resulting in changes in the electricity demand of users. The changes in electricity demand will have a great impact on the user's demand electricity bill. However, in many cases, the income generated by response invitations is far from enough to cover the increased electricity bill due to the increase in user demand.

[0007] In this case, users cannot obtain economic benefits from responding to invitations, and may even face economic losses due to participating in responding to invitations. This greatly affects users' enthusiasm for responding to dispatching invitations, resulting in more and more users being unwilling to participate in demand response invitations, thereby reducing the participation group of demand response invitations. The reduction in the user load capacity of those participating in peak regulation and frequency modulation in turn reduces the effectiveness of the virtual power plant, and it cannot fully play its role in balancing the load fluctuations on the power generation side and improving the flexible regulation ability of the power system, forming a vicious cycle. Summary of the Invention

[0008] The present invention aims to provide a regulation method for demand-side response dispatching invitations considering the maximum demand, so as to solve the technical problems of low efficiency of the virtual power plant in responding to invitations and poor ability to balance load fluctuations on the power generation side.

[0009] To achieve the above object, the present invention adopts the following technical solution: A regulation method for demand-side response dispatching invitations considering the maximum demand, comprising the following steps:

[0010] First step, obtain the data affecting the user load prediction;

[0011] Second step, input the data affecting the user load prediction into the user load prediction model to obtain the predicted curve of the user's electricity consumption load for the next 2 days;

[0012] Third step, collect the energy storage device information of the user according to the user ID signed for the demand-side response;

[0013] Fourth step, obtain the proposed demand response invitation information from the power grid dispatching center;

[0014] Fifth step, calculate the maximum demand of the user according to the predicted curve of the user's electricity consumption load and the actual curve of the user's electricity consumption load;

[0015] Sixth step, calculate the output curve of the user's energy storage device during the response period according to the demand response invitation information and the maximum demand, and then calculate the user response ability curve during this period. Adjust the demand response invitation information according to the output curve of the user's energy storage device and the user response ability curve, and conduct response declarations with the adjusted demand response invitation;

[0016] Seventh step, during the invitation response period, monitor the actual electricity consumption load of the user in real time, and regulate the output of the energy storage device to ensure that the maximum demand limit is not exceeded, and complete the demand-side response invitation.

[0017] The advantages of this solution are:

[0018] One, break the conventional thinking and safeguard the economic interests of users

[0019] Due to the common thinking of virtual power plants that prioritize the interests of the power grid, when the invitation response efficiency is low, those skilled in the art tend to increase the electricity consumption on the user side or raise the electricity price incentives to attract users. However, this approach often fails to truly enhance the economic benefits of users. For example, when implementing demand response in a certain region, simply increasing the incentive electricity price, users increase their electricity load to obtain subsidies, but the resulting substantial increase in demand charges far exceeds the incentive income, and the actual economic burden of users increases.

[0020] This solution breaks out of this mindset and fully considers the economic interests of users while ensuring the grid regulation requirements. Through reasonable planning and control, it ensures that users will not bear excessive additional costs due to the increase in electricity demand when participating in demand response, truly achieving a win-win situation for users and the grid.

[0021] II. Balancing User Electricity Demand and Grid Regulation

[0022] Those skilled in the art usually believe that considering the user's electricity demand will limit the user's load regulation ability when responding to invitations, affecting the invitation response efficiency, and consider it contrary to improving the grid regulation ability. But in fact, without considering the user's electricity demand, users may refuse to participate in the response due to concerns about increased electricity bills, which instead reduces the grid regulation ability.

[0023] This solution cleverly finds a balance between the two. Through precise calculation and reasonable arrangement, it gives full play to the user's load regulation ability on the premise of meeting the user's electricity demand. For example, according to the user's historical electricity consumption data and real-time load conditions, personalized response strategies are formulated to not only ensure the normal electricity use of users but also provide effective regulation resources for the grid.

[0024] III. Advance Planning to Solve Scheduling Problems

[0025] Those skilled in the art believe that dispatchers can only hurriedly formulate scheduling strategies that meet the maximum demand limit based on real-time and uncertain load conditions at the moment of invitation response. The decision-making time is urgent and faces many uncertainties, making it difficult to achieve effective demand limit scheduling.

[0026] This solution provides sufficient time for dispatchers to plan by predicting the user's load conditions in advance. Dispatchers can plan the charge and discharge curves of flexible loads or energy storage devices during the invitation period in advance, just like drawing a "battle map" in advance. During the invitation response period, only fine-tuning according to the actual situation is needed, without having to make decisions from scratch, greatly reducing the pressure and difficulty of real-time decision-making and enabling the effective implementation of scheduling strategies considering the user's maximum demand.

[0027] IV. Demand Control Enhances User Participation Enthusiasm

[0028] Through long-term accumulation of work experience, those skilled in the art creatively discovered the impact of demand control on the efficiency of the invitation effect. If the demand is not controlled, exceeding the maximum demand will bring additional costs to the responding users, resulting in users being less active in responding to invitations or even completely not participating in demand-side response, greatly reducing the group of participants in the invitation response.

[0029] This solution eliminates the users' concerns by strictly controlling the maximum demand. When participating in the response, users do not need to worry about high electricity bills due to exceeding the demand limit, so they are more willing to actively respond to invitations. This not only enhances the dispatching ability of the power grid's demand-side response but also promotes the healthy development of the demand-side response market.

[0030] V. Advance Planning to Reduce the Pressure of Real-Time Decision-Making

[0031] By predicting the user load situation in advance, dispatchers can plan the charge and discharge curves of flexible loads or energy storage devices during the invitation period in advance. During the invitation response period, dispatchers only need to execute according to the pre-established plan and make fine-tuning according to the real-time load situation. This way of advance planning avoids the situation of making hasty decisions during the invitation response, greatly reducing the pressure and difficulty of real-time decision-making and making the dispatching work more efficient and orderly.

[0032] This solution optimizes the invitation response strategy by comprehensively considering the user's maximum demand limit, improving the enthusiasm of users to participate in the invitation response, thus increasing the group of participants in the demand response invitation, enabling the power grid to dispatch more user load resources, and effectively improving the effectiveness of the invitation response.

[0033] By giving full play to the frequency modulation and peak shaving functions of the user-side resources, this solution can balance the load fluctuations on the power generation side and improve the stability and reliability of the power system. When dealing with sudden changes in power demand or fluctuations in new energy generation, the power grid can adjust more calmly to ensure the continuity and stability of power supply.

[0034] With the increasing access of a high proportion of new energy, the pressure on the power grid to absorb and utilize it is growing. This solution promotes the absorption and utilization of new energy and the optimization and upgrading of the energy structure by guiding users to increase electricity consumption when new energy generation is sufficient and reduce electricity consumption when generation is insufficient through demand-side response. For power grid enterprises, this solution reduces the demand for reserve capacity and operating costs by optimizing the dispatching strategy. At the same time, users' participation in demand-side response can also reduce their own electricity costs to a certain extent, achieving a win-win situation for the power grid and users.

[0035] Preferably, as an improvement, the energy storage device information includes energy storage capacity, charging power, and discharging power; the demand response invitation information includes response type, response start time, response end time, and the user load baseline during the response period; the response type includes valley filling response and peak shaving response;

[0036] The output situation of the user's energy storage device during this response period is calculated using constraint equations. When the response type is valley filling response, the energy storage device needs to be charged during this period, and its constraint equations are as follows:

[0037]

[0038] When the response type is peak shaving response, the energy storage device needs to discharge during this period, and its constraint equations are as follows:

[0039]

[0040] Among them, C t is the charging power or discharging power of the energy storage device at this moment, C max is the maximum charging power or maximum discharging power of the energy storage device, P t is the predicted user power consumption load at this moment, M is the user's maximum demand, and Q is the total capacity of the energy storage device.

[0041] The beneficial effects of this improvement are as follows: By considering the specific parameters of the energy storage device, such as energy storage capacity, charging power, and discharging power, the actual ability of the energy storage device in demand response can be evaluated more accurately. And because the charging and discharging ramping rate of the energy storage is relatively fast, climbing from 0 to the maximum in no more than 1 minute, the ramping rate problem of the energy storage is not considered when calculating the charging and discharging of the energy storage, so as to simplify the algorithm and reduce the implementation difficulty. Different constraint conditions are formulated according to different response types, such as valley filling response and peak shaving response, making the response invitation more in line with the actual needs of the power grid and improving the pertinence and effectiveness of the response.

[0042] Preferably, as an improvement, according to the output curve of the energy storage device and the predicted curve of the user power consumption load, the user response ability curve is calculated, and the calculation formula is as follows:

[0043] E t =P t -C t -Pb t ;

[0044] Among them, E t is the response ability of the user at this moment, P t is the predicted user power consumption load at this moment, C t is the output power of the energy storage device at this moment, and Pb t is the user's baseline power consumption load at this moment.

[0045] The beneficial effects of this improvement are as follows: By comprehensively considering the actual output capacity of the energy storage device and the predicted situation of the user's electricity load, the actual response ability of the user in demand response can be more accurately evaluated. This evaluation method takes into account the dynamic characteristics of the energy storage device and the diversity of the user's electricity consumption behavior, making the evaluation results closer to the actual situation. Based on the calculation results of the user response ability curve, the power grid dispatching center can formulate more optimized demand response strategies. For example, during peak load periods, users with strong response ability and large energy storage device output can be preferentially selected to participate in peak shaving response, thereby more effectively balancing the power grid load.

[0046] Preferably, as an improvement, the user load prediction influencing data includes the user's historical electricity consumption data, weather data of the user's location, and holiday data. The weather data includes historical temperature, future predicted temperature, historical humidity, and future predicted humidity; the holiday data includes the dates of historical and future holidays.

[0047] The beneficial effects of this improvement are as follows: The historical electricity consumption data provides basic information on the user's electricity consumption habits. Weather data, such as temperature and humidity, reflects the impact of the external environment on the user's electricity demand, while holiday data captures the changes in the user's electricity consumption pattern on special dates. The comprehensive consideration of these factors can significantly improve the accuracy of user load prediction. The introduction of multiple types of data enables the model to better adapt to the changes in electricity demand of different users, different regions, and different time periods, improving the generalization ability and adaptability of the model.

[0048] Preferably, as an improvement, a deep autoregressive recurrent neural network is used to construct the user load prediction model. The input layer of the user load prediction model is used to receive the user's historical 7-day electricity load data, the historical 7-day weather data of the location, the future 2-day weather prediction data of the location, and the holiday situations of the location in the historical 7 days and the future 2 days;

[0049] The hidden layer of the user load prediction model uses two recurrent neural network layers, with 40 neurons in each layer, and uses an activation function for non-linear transformation;

[0050] The output layer of the user load prediction model is used to output the predicted user's future 2-day electricity load prediction curve.

[0051] The beneficial effects of this improvement are as follows: The deep autoregressive recurrent neural network is good at processing time series data and can capture the changing rules of the user's electricity load over time. By introducing recurrent neural network layers, the model can better understand the relationship between historical data and future predictions.

[0052] The design of two recurrent neural network layers with 40 neurons in each layer increases the depth and width of the model, enabling the model to learn more complex non-linear relationships and improving the model's expressive ability. By outputting the predicted curve of electricity load for the next two days, the model can provide longer-term prediction information for power grid scheduling and energy storage device planning, which helps to formulate more reasonable demand response strategies.

[0053] Preferably, as an improvement, the user load prediction model is set with hyperparameters of a step size of 96*2 and a dropout rate of 0.1.

[0054] The beneficial effects of this improvement are as follows: A step size of 96*2 means that the model considers a longer time window during prediction. Assuming each step represents a certain time interval, such as 15 minutes, then 96 steps represent one day, and 96*2 represents two days. This helps the model capture the changing trend of electricity load over a longer period and improves the stability of prediction. Dropout is a regularization technique that randomly discards a part of neurons during training to prevent the model from overfitting. Setting the dropout rate to 0.1 means that in each training iteration, 10% of the neurons are randomly discarded, which helps to reduce the complexity of the model and improve the generalization ability of the model.

[0055] Preferably, as an improvement, in the fifth step, the maximum value of the average power per 15 minutes in the user's electricity consumption historical data for the current month is used as the user's maximum demand. The generated predicted curve of the user's electricity load is compared with the maximum demand value to determine whether the value of the predicted curve exceeds the maximum demand value at any time point. If there is a breakthrough point, the largest predicted value among the breakthrough points is used as the new user's maximum demand.

[0056] The beneficial effects of this improvement are as follows: By directly using the maximum value of the average power per 15 minutes in the user's electricity consumption historical data for the current month as the user's maximum demand, rather than based on the average value of partial months or historical data, it can more accurately reflect the user's current electricity consumption habits and load characteristics. When a breakthrough point is predicted, using the largest predicted value among the breakthrough points as the new user's maximum demand helps to formulate more effective demand response strategies. The power grid dispatching center can adjust the charging and discharging plans of energy storage devices and the electricity consumption plans of users in advance according to the new predicted maximum demand value to ensure that the user's electricity load does not exceed this new predicted value during the demand response period, thus ensuring the stable operation of the power grid and the economic interests of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flowchart of an embodiment of the present invention.

[0058] Figure 2 It is a bar chart of the response calculation time consumption of the virtual power plant of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0059] The following is a further detailed description through specific embodiments:

[0060] Embodiment

[0061] Basically as shown in the appendix Figure 1 A regulation method for a demand - side response scheduling invitation considering the maximum demand includes:

[0062] First step: Obtain user load prediction influence data, which includes user historical electricity consumption data, weather data of the user's location, and holiday data. The weather data includes historical temperature, future predicted temperature, historical humidity, and future predicted humidity. The holiday data includes the dates of historical and future holidays.

[0063] Second step: Input the user load prediction influence data into the user load prediction model to obtain the electricity load prediction curve of the user for the next 2 days.

[0064] The Deep Autoregressive Recurrent Neural Network (DAR - RNN) is used to model the user load prediction model. DAR - RNN is a model that combines the advantages of deep learning and recurrent neural network (RNN), and is particularly suitable for processing time - series data.

[0065] And set hyperparameters of step size 96*2, number of layers 2, 40 neurons in the hidden layer, and dropout 0.1, and perform modeling training on the load prediction influence data of each demand - side response signed user. By setting this way, one day is divided into 96 time steps, such as one time step every 15 minutes, and the load data for the next two days is predicted. And this user load prediction model contains two recurrent neural network layers, and each hidden layer contains 40 neurons. The hyperparameter of 0.1 is used to prevent the model from overfitting, which is achieved by randomly discarding the outputs of some neurons during the training process.

[0066] The input layer of this user load prediction model is used to receive the user's historical 7 - day electricity load data, the historical 7 - day weather data of the location, the future 2 - day weather prediction data of the location, and the holiday situations of the location for the historical 7 days and the next 2 days; the hidden layer uses two recurrent neural network layers, with 40 neurons in each layer, and uses an activation function (such as ReLU) for non - linear transformation; the output layer of the user load prediction model is used to output the predicted electricity load prediction curve of the user for the next 2 days.

[0067] Step 3: According to the user ID signed up for demand response, collect the energy storage device information of this user. The energy storage device information includes energy storage capacity, charging power, and discharging power. Since the charging and discharging ramping rates of energy storage are relatively fast, ramping up from 0 to the maximum within no more than 1 minute, the ramping rate issue of energy storage is not considered when calculating energy storage charging and discharging to simplify the algorithm.

[0068] Step 4: Obtain the issued demand response invitation information from the power grid dispatching center. The demand response invitation information includes response type, response start time, response end time, and the user load baseline during the response period. The response type includes valley filling response and peak shaving response. The user load baseline during the response period is the electricity load situation of the user when not participating in the response during the demand response period.

[0069] Step 5: Calculate the maximum demand of the user based on the user's predicted electricity load curve and the actual electricity load curve of the user.

[0070] The user's maximum demand is the maximum value of the average power of this user every 15 minutes within a certain settlement cycle. Since it has not reached the end of the month at this time, the actual maximum demand of the user cannot be directly counted. That is, the maximum demand for electricity billing is calculated based on the maximum power of the average power of this user every 15 minutes this month. Therefore, it is necessary to count a temporary maximum demand as a reference according to the user's electricity consumption history this month. This value represents the maximum power of the average power of the user every 15 minutes within this month so far.

[0071] Compare the generated user predicted electricity load curve with the temporary maximum demand value to determine whether the value of the predicted curve exceeds the temporary maximum demand value at any time point. If there is a breakthrough point, use the predicted value at the breakthrough point as the new maximum demand. If the predicted curve exceeds the temporary maximum demand at multiple time points, then the largest one of them needs to be selected as the new maximum demand.

[0072] Step 6: Calculate the output of the user's energy storage device during this response period according to the demand response invitation information and the maximum demand, and calculate the invitation response ability of the user at this moment. Adjust the demand response invitation based on this invitation response ability and the output of the user's energy storage device, and submit a response declaration with the adjusted demand response invitation; the demand response invitation contains the invitation content of the user's response degree.

[0073] When the response type is valley filling response, the energy storage device needs to be charged during this period, and its constraint equations are as follows:

[0074]

[0075] When the response type is peak shaving response, the energy storage device needs to be discharged during this period, and its constraint equations are as follows:

[0076]

[0077] Among them, C t is the charging power or discharging power of the energy storage device at this moment, and C max is the maximum charging power or maximum discharging power of the energy storage device, P t is the predicted user power consumption load at this moment, M is the user's maximum demand, and Q is the total capacity of the energy storage device.

[0078] By solving the linear equations, the output power of the energy storage device at each moment is obtained, that is, C t . Then, according to the output curve of the energy storage device and the predicted curve of the user power consumption load, the user response ability curve is calculated. The calculation formula is as follows:

[0079] E t = P t - C t - Pb t ;

[0080] Among them, E t is the response ability of the user at this moment, P t is the predicted user power consumption load at this moment, C t is the output power of the energy storage device at this moment, and Pb t is the user's baseline power consumption load at this moment.

[0081] Step 7: During the invitation response period, monitor the actual power consumption load of the user in real time, and regulate the output of the energy storage device to ensure that the maximum demand limit is not exceeded and the demand-side response invitation is completed.

[0082] This way of real-time monitoring and flexible regulation can, for users, not only effectively control electricity costs, avoid additional costs caused by exceeding the demand, but also improve energy utilization efficiency and achieve energy conservation and emission reduction. For the power grid company, it helps to maintain the stable operation of the power grid, reduce the power grid load fluctuation, and improve the reliability and security of power supply.

[0083] In this embodiment, by predicting the user load situation in advance, sufficient time is provided for the dispatcher to plan. The dispatcher can plan the charge and discharge curves of the flexible load or the energy storage device during the invitation period in advance, just like drawing a "battle map" in advance. During the invitation response period, only fine-tuning according to the actual situation is required, without having to make decisions from scratch, greatly reducing the pressure and difficulty of real-time decision-making.

[0084] As shown in the appendix Figure 2As shown, a bar chart of the time-consuming for the virtual power plant response calculation is provided to reflect the advantages of this solution in terms of calculation efficiency. As the number of energy storage devices increases, the time-consuming for the virtual power plant response calculation also increases. However, by comparing the results of manual calculation and this embodiment, it can be clearly seen that this embodiment is more efficient in terms of calculation time-consuming. For example, when the number of energy storage devices is 1, manual calculation takes 7 minutes, while the solution of this example only takes 3.5 minutes; when the number of energy storage devices increases to 3, manual calculation takes 22 minutes, while this embodiment only takes 4 minutes. This intuitively demonstrates the advantages of the solution of this example in terms of calculation efficiency. Even in the face of more energy storage devices and more complex calculation scenarios, the solution of this example can still maintain a relatively high calculation speed. This helps to make decisions faster in practical applications and improve the response speed and overall operation efficiency of the virtual power plant.

[0085] This technical solution not only breaks the conventional thinking, safeguards the economic interests of users, but also takes into account the electricity demand of users and the grid regulation. It also resolves the dispatching problem through advance planning. And the Figure 2 situation of the time-consuming for the virtual power plant response calculation shown with the increase in the number of energy storage devices further proves the advantages of this solution in terms of calculation efficiency, enabling the implementation of the response demand regulation considering the electricity demand of users.

[0086] The above are only the embodiments of the present invention. Specific technical solutions and / or common knowledge such as characteristics well known in the art are not described in detail here. It should be noted that for those skilled in the art, without departing from the technical solution of the present invention, several deformations and improvements can still be made, and these should also be regarded as the protection scope of the present invention, which will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners described in the specification can be used to interpret the content of the claims.

Claims

1. A method for adjusting the demand-side response scheduling invitation considering the maximum demand, characterized in that: The following steps are involved: The first step is to obtain user load forecast impact data; The second step is to input the user load forecast impact data into the user load forecast model to obtain the user's power load forecast curve for the next two days; Step 3: Collect the energy storage device information of the user according to the user ID of the demand side response contract; Step 4: The power grid dispatching center obtains the proposed demand response invitation information; Step 5: Calculate the user's maximum demand based on the user's power load forecast curve and the user's power load actual curve; Step 6: Calculate the output curve of the user's energy storage device in the response period according to the demand response invitation information and the maximum demand, and then calculate the user's response capacity curve in the period, adjust the demand response invitation information according to the user's energy storage device output curve and the user's corresponding capacity curve, and make a response declaration according to the adjusted demand response invitation; Step 7. During the invitation response period, monitor the user's actual power load in real time, adjust the output of the energy storage equipment, ensure that the maximum demand limit is not exceeded, and complete the demand-side response invitation.

2. The method for adjusting the demand-side response scheduling invitation considering the maximum demand according to claim 1, characterized in that: The energy storage device information includes energy storage capacity, charging power and discharging power; the demand response invitation information includes response type, response start time, response end time and response period user load baseline; response type includes valley filling response and peak shaving response; The constraint equation is used to calculate the output of the user's energy storage equipment during the response period. When the response type is valley filling response, the energy storage equipment needs to be charged during this period, and the constraint equation group is as follows: When the response type is peak shaving response, the energy storage device needs to discharge during this period, and its constraint equations are as follows: Among them, C t is the charging power or discharging power of the energy storage device at that moment, C max is the maximum charging power or maximum discharging power of the energy storage device, P t is the user's power load predicted at that moment, M is the user's maximum demand, and Q is the total capacity of the energy storage device.

3. The method for adjusting the demand-side response scheduling invitation considering the maximum demand according to claim 2, characterized in that: According to the output curve of the energy storage device and the user's power load forecast curve, the user response capability curve is calculated. The calculation formula is as follows: E t =P t -C t -Pb t ; Among them, E t is the user’s responsiveness at that moment, P t is the user power load predicted at that moment, C t is the output power of the energy storage device at that moment, Pb t It is the user's baseline power load at that moment.

4. The method for adjusting the demand-side response scheduling invitation considering the maximum demand according to claim 1, characterized in that: The user load forecast influencing data includes the user's historical electricity consumption data, weather data and holiday data of the user's location. The weather data includes historical temperature, future predicted temperature, historical humidity and future predicted humidity; the holiday data includes the dates of historical and future holidays.

5. The method for adjusting the demand-side response scheduling invitation considering the maximum demand according to claim 1, characterized in that: The user load prediction model is constructed by using a deep autoregressive recurrent neural network. The input layer of the user load prediction model is used to receive the user's historical 7-day electricity load data, the location's historical 7-day weather data, the location's weather forecast data for the next 2 days, and the location's historical 7-day and next 2-day holiday conditions; The hidden layer of the user load prediction model uses two recurrent neural network layers with 40 neurons in each layer and uses activation functions for nonlinear transformation; The output layer of the user load forecasting model is used to output the predicted user's electricity load forecast curve for the next two days.

6. The method for adjusting the demand-side response scheduling invitation considering the maximum demand according to claim 5, characterized in that: The user load prediction model sets the step size to 96*2 and the hyper parameters of random dropout to 0.

1.

7. The method for adjusting the demand-side response scheduling invitation considering the maximum demand according to claim 6, characterized in that: In the fifth step, the maximum value of the average power every 15 minutes in the user's electricity consumption history data for the month is taken as the user's maximum demand, and the generated user electricity load forecast curve is compared with the maximum demand value to determine whether the value of the forecast curve exceeds the maximum demand value at any time point. If there is a breakthrough point, the maximum forecast value at the breakthrough point is taken as the new user maximum demand.

Images