Intelligent electric energy dispatching method and system

Abstract

The present invention relates to an intelligent electric energy dispatching method and system, the method comprising the following steps: obtaining user-side load demand and electricity price information of the current time period; preprocessing the load demand and electricity price information, predicting the load demand of the future time period through the LSTM network according to the load demand, and obtaining the predicted load demand; predicting the future electricity price change through time series analysis according to the electricity price information, and obtaining the predicted electricity price; setting the optimization target, obtaining the energy distribution scheme according to the user-side load demand and the electricity price information of the current time period through the particle swarm optimization algorithm, and obtaining the dispatching scheme according to the predicted load demand and the predicted electricity price through the MPC algorithm; adjusting the energy storage state of the user side in real time according to the energy distribution scheme, adjusting the power transmission of the user side in real time according to the dispatching scheme, and completing the electric energy dispatching. Compared with the prior art, the present invention realizes efficient electric energy management and reduces the operating cost.

Description

Technical Field

[0001] The present invention relates to the technical field of voltage stabilization, and in particular to an intelligent electric energy dispatching method and system. Background Art

[0002] With the rapid development of power systems, especially in the widespread application of distributed energy, flexible DC technology and smart grids, user-side power management and voltage stabilization technology have become increasingly important. In traditional voltage stabilization solutions, AC power is usually converted to DC through a rectifier, and then the DC is converted back to AC through an inverter, and finally the voltage is stabilized and output through a voltage regulator controller. Although this technical solution can achieve basic voltage stabilization functions, it has some obvious shortcomings when facing complex load demands and fluctuating electricity prices, making it difficult to achieve efficient energy management and optimization.

[0003] At the same time, as the power market gradually introduces time-of-use electricity prices and more dynamic load changes, traditional voltage regulators and control systems face increasing challenges, especially in terms of energy scheduling and economy. Therefore, how to optimize the application of flexible DC technology and introduce intelligent energy management systems has become the key to improving system efficiency and reducing operating costs. Existing flexible DC voltage stabilization solutions have multiple technical defects. First, the system mainly focuses on voltage stability and ignores energy economic scheduling, especially in the time-of-use electricity price environment, and cannot flexibly respond to electricity price fluctuations, resulting in low energy utilization efficiency and high operating costs. Secondly, traditional PI / PID controllers are prone to response lag or over-regulation when the load changes rapidly, affecting system stability. Existing solutions are also difficult to perform intelligent load regulation, cannot cope with complex load changes, and reduce system efficiency. In addition, the existing system structure is relatively complex, which increases manufacturing costs and maintenance difficulties, and is prone to system failures. In summary, the existing technology has obvious limitations in energy scheduling, load response and system complexity, and cannot meet the needs of modern power systems for efficient, economical and intelligent power management. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent power dispatching method and system in order to overcome the defects of the above-mentioned prior art, thereby achieving efficient power management and reducing operating costs.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] An intelligent electric energy dispatching method comprises the following steps:

[0007] Obtain user-side load demand and electricity price information for the current time period;

[0008] Preprocessing the load demand and electricity price information, predicting the load demand in a future time period through an LSTM network according to the load demand, and obtaining a predicted load demand; predicting future electricity price changes through time series analysis according to the electricity price information, and obtaining a predicted electricity price;

[0009] Setting an optimization goal, obtaining an energy allocation plan based on the user-side load demand and the electricity price information of the current time period through a particle swarm optimization algorithm, and obtaining a scheduling plan based on the predicted load demand and the predicted electricity price through an MPC algorithm;

[0010] The power output of the user-side device is adjusted in real time according to the energy allocation scheme, and the voltage value of the user-side is adjusted in real time according to the scheduling scheme to complete the power scheduling.

[0011] Furthermore, the optimization objectives include economic objectives, power balance objectives and operational reliability objectives. In the process of obtaining the energy allocation plan according to the user-side load demand and the electricity price information of the current time period through the particle swarm optimization algorithm, the power output of the user-side equipment is adjusted in real time according to the economic objective, and the power balance objective and the operational reliability objective are used as constraints; in the process of obtaining the scheduling plan according to the predicted load demand and the predicted electricity price through the MPC algorithm, the voltage value on the user side is adjusted in real time according to the economic objective, and the power balance objective and the operational reliability objective are used as constraints.

[0012] Furthermore, the operational reliability target is:

[0013] ,

[0014] In the formula, is the voltage value on the user side during period t, is the current value on the user side during period t, is the output power of the user side during period t, is the safe lower limit of voltage value, is the safety upper limit of the voltage value, is the safe lower limit of the current value, is the safe upper limit of the current value, is the maximum power limit.

[0015] Furthermore, the objective function of the economic objective is:

[0016] ,

[0017] In the formula, is the electricity price in period t, is the power value on the user side during period t. The power value on the user side includes output power, load demand power and energy storage power. The total time frame for optimization.

[0018] Furthermore, the power balance target is:

[0019] ,

[0020] In the formula, is the output power of the user side in period t, is the load demand power during period t, is the energy storage power in period t, is the power loss during period t.

[0021] Furthermore, the output power is:

[0022] ,

[0023] In the formula, is the maximum output power, is the average electricity price, is the electricity price in period t, is the adjustment coefficient;

[0024] The calculation formula of the energy storage power is:

[0025] ,

[0026] In the formula, is the energy storage power in period t, is the highest electricity price during the operation period, is the energy storage regulation coefficient.

[0027] Furthermore, the future load demand is predicted by the LSTM network model, and the prediction equation of the load demand is:

[0028] ,

[0029] In the formula, For the future The load demand forecast value at the time, is the mapping of the LSTM network model, It is the load demand data at different times in the past.

[0030] Furthermore, the objective function of the particle swarm optimization algorithm is:

[0031] ,

[0032] In the formula, is the unit energy consumption output of the i-th device on the user side, is the power output of the ith device on the user side, is the total power output at the user side, is the power required by the user side, is the penalty coefficient for power balancing.

[0033] Furthermore, the objective function of the MPC algorithm is:

[0034] ,

[0035] In the formula, is the control input on the user side, To predict the duration, is the penalty factor for voltage deviation, is the set voltage value, is the electricity price in period t, is the power value on the user side during period t, is the voltage value at the user side during period t.

[0036] According to another aspect of the present invention, there is provided an intelligent electric energy dispatching system, comprising:

[0037] A data acquisition module is used to obtain user-side load demand and electricity price information for the current time period;

[0038] A data preprocessing module is used to preprocess the load demand and electricity price information, predict the load demand in the future time period through the LSTM network according to the load demand, and obtain the predicted load demand; predict the future electricity price changes through time series analysis according to the electricity price information, and obtain the predicted electricity price;

[0039] An optimization strategy generation module is used to set optimization goals, obtain the optimal energy allocation plan according to the user-side load demand and the electricity price information of the current time period through a particle swarm optimization algorithm, and obtain the optimal scheduling plan according to the predicted load demand and predicted electricity price through an MPC algorithm;

[0040] The scheduling execution module is used to adjust the power output of the user-side equipment in real time according to the energy allocation plan, and adjust the voltage value of the user-side in real time according to the scheduling plan to complete the power scheduling.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. The present invention predicts the load demand in the future time period through the LSTM network according to the user-side load demand and the electricity price information in the current time period to obtain the predicted load demand, and predicts the future electricity price changes through time series analysis to obtain the predicted electricity price; according to the optimization target, the particle swarm optimization algorithm is used to obtain the optimal energy allocation according to the user-side load demand and the electricity price information in the current time period, and the MPC algorithm is used to obtain the optimal scheduling plan according to the predicted load demand and the predicted electricity price; according to the optimal energy allocation and the optimal scheduling plan, the source-side energy storage state and power transmission are adjusted in real time, thereby realizing efficient energy management and improving the efficiency of power management.

[0043] 2. The present invention sets optimization goals, including economic goals, power balance goals and operational reliability goals, to ensure power balance between the source side and the user side while meeting load demand, thereby ensuring that the voltage and current of the flexible DC line remain within a safe operating range, and optimizing the storage and release timing of electric energy by utilizing changes in time-of-use electricity prices, reducing energy supply during high-price electricity periods, and storing energy during low-price electricity periods, thereby significantly reducing electricity costs and optimizing economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a flow chart of the intelligent electric energy dispatching method proposed in the present invention. DETAILED DESCRIPTION

[0045] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0046] Abbreviations involved:

[0047] Long Short-Term Memory Network: Long Short-Term Memory, LSTM

[0048] Model Predictive Control: Model Predictive Control, MPC.

[0049] Example 1

[0050] This embodiment provides an intelligent power dispatching method, such as Figure 1 As shown, the following steps are included.

[0051] S1. Obtain user-side load demand and electricity price information for the current time period.

[0052] Collect the power demand of user-side loads in real time, including active power and reactive power demand data. By connecting with the power market, obtain the electricity price information of the current time period in real time and predict the trend of electricity price changes in the future.

[0053] S2. Preprocess the load demand and electricity price information, and predict the load demand in the future time period through the LSTM network according to the load demand to obtain the predicted load demand; predict the future electricity price changes through time series analysis according to the electricity price information to obtain the predicted electricity price.

[0054] Pre-process the collected load demand and electricity price data to remove abnormal data and ensure the accuracy of the data.

[0055] The future load demand is predicted by the LSTM network model. The load demand prediction equation is:

[0056] ,

[0057] In the formula, For the future The load demand forecast value at the time, is the mapping of the LSTM network model, It is the load demand data at different times in the past.

[0058] The time series method used in this embodiment is the autoregressive integrated moving average model (ARIMA). The ARIMA model can effectively capture the law of electricity price fluctuations and predict future electricity price trends by analyzing the autocorrelation and seasonal changes in historical electricity price data. This method can provide accurate electricity price forecasts for energy scheduling algorithms, helping the system to store energy during low electricity price periods and reduce energy consumption during high electricity price periods, thereby optimizing the economic efficiency of power scheduling. The model equation is:

[0059] ,

[0060] In the formula, is the electricity price at time t, c is the long-term average level of electricity price, is the autoregressive coefficient, which indicates the relationship between the current electricity price and the electricity price at p time points in the past. is white noise, indicating the random error at time t, is the sliding average coefficient, which represents the relationship between the current error and the errors at the past q time points.

[0061] S3. Set the optimization goal, use the particle swarm optimization algorithm to obtain the energy allocation plan based on the user-side load demand and the electricity price information of the current time period, and use the MPC algorithm to obtain the scheduling plan based on the predicted load demand and predicted electricity price.

[0062] The optimization objectives include economic objectives, power balance objectives, and operational reliability objectives. The economic objective is to minimize the user's electricity cost; the power balance objective is to ensure the power balance on the user side while meeting the load demand; the operational reliability objective is to ensure that the voltage and current of the flexible DC line remain within the safe operating range. In the process of obtaining the energy allocation plan based on the user's load demand and the electricity price information of the current time period through the particle swarm optimization algorithm, the power output of the user-side equipment is adjusted in real time according to the economic objective, and the power balance objective and operational reliability objective are used as constraints; in the process of obtaining the scheduling plan based on the predicted load demand and predicted electricity price through the MPC algorithm, the voltage value on the user side is adjusted in real time according to the economic objective, and the power balance objective and operational reliability objective are used as constraints.

[0063] The operational reliability objectives are:

[0064] ,

[0065] In the formula, is the voltage value on the user side during period t, is the current value on the user side during period t, is the output power of the user side during period t, is the safe lower limit of voltage value, is the safety upper limit of the voltage value, is the safe lower limit of the current value, is the safe upper limit of the current value, is the maximum power limit.

[0066] The objective function of the economic goal is:

[0067] ,

[0068] In the formula, is the electricity price in period t, is the power value on the user side during period t. The power value on the user side includes output power, load demand power and energy storage power. The total time frame for optimization.

[0069] The power balance objectives are:

[0070] ,

[0071] In the formula, is the output power of the user side in period t, is the load demand power during period t, is the energy storage power in period t, is the power loss during period t;

[0072] Among them, the output power is:

[0073] ,

[0074] In the formula, is the maximum output power, is the average electricity price, is the electricity price in period t, is the adjustment coefficient.

[0075] The calculation formula of energy storage power is:

[0076] ,

[0077] In the formula, is the energy storage power in period t, is the highest electricity price during the operation period, is the electricity price in period t, is the energy storage regulation coefficient.

[0078] The particle swarm optimization algorithm is used to obtain the energy allocation plan based on the user-side load demand and the electricity price information of the current time period. The particle swarm optimization algorithm optimizes the system's energy scheduling strategy by simulating the movement and search process of particles, and finds the optimal solution that minimizes cost or satisfies power balance. The objective function of the particle swarm optimization algorithm is:

[0079] ,

[0080] In the formula, is the unit energy consumption output of the i-th device on the user side, is the power output of the ith device on the user side, is the total power output at the user side, is the power required by the user side, is the penalty coefficient for power balancing.

[0081] The MPC algorithm is used to obtain a dispatching plan based on the predicted load demand and predicted electricity price. The dispatching plan includes energy transmission amount and transmission time. The objective function of the MPC algorithm is:

[0082] ,

[0083] In the formula, is the control input on the user side, To predict the duration, is the penalty factor for voltage deviation, is the set voltage value, is the electricity price in period t, is the power value on the user side during period t, is the voltage value at the user side during period t.

[0084] S4. Adjust the power output of the user-side equipment in real time according to the energy allocation plan, and adjust the voltage value on the user side in real time according to the scheduling plan to complete the power scheduling.

[0085] The energy allocation plan includes optimizing the power output of each device, while ensuring that the total power meets the load demand and minimizes the total energy cost. The scheduling plan includes voltage control on the user side. The scheduling plan meets the economic goals and aims to achieve voltage stability while minimizing costs and improving energy efficiency.

[0086] In order to execute the scheduling scheme more accurately, this embodiment considers the power transmission efficiency and delay in each time period, and the formula is as follows:

[0087] ,

[0088] In the formula, is the actual transmitted power, The transmission efficiency coefficient varies with system load and equipment status. By introducing the transmission efficiency coefficient, the loss and delay in the actual power transmission process can be more accurately reflected, ensuring that the execution of the scheduling plan is more in line with the actual operation situation.

[0089] In another preferred embodiment, the steps are further included:

[0090] S5. Collect the operating data of the flexible DC line and the user side in real time, and adjust and optimize the dispatching strategy based on the operating data.

[0091] Collect the operation data of the flexible DC line and the user side in real time, including voltage, current and load changes. Adjust and optimize the dispatching strategy based on the feedback data to ensure stability and economy when the electricity price fluctuates and the load changes. Through continuous data collection, optimization calculation and feedback adjustment, dynamic optimization of energy management is achieved.

[0092] Example 2

[0093] This embodiment provides an intelligent electric energy dispatching system, including:

[0094] A data acquisition module is used to obtain user-side load demand and electricity price information for the current time period;

[0095] The data preprocessing module is used to preprocess the load demand and electricity price information, and predict the load demand in the future time period through the LSTM network according to the load demand to obtain the predicted load demand; and predict the future electricity price changes through time series analysis according to the electricity price information to obtain the predicted electricity price;

[0096] The optimization strategy generation module is used to set the optimization goal, obtain the optimal energy allocation plan based on the user-side load demand and the electricity price information of the current time period through the particle swarm optimization algorithm, and obtain the optimal scheduling plan based on the predicted load demand and predicted electricity price through the MPC algorithm;

[0097] The scheduling execution module is used to adjust the power output of the user-side equipment in real time according to the energy allocation plan, and adjust the voltage value on the user side in real time according to the scheduling plan to complete the power scheduling.

[0098] The rest is the same as in Example 1.

[0099] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. An intelligent electric energy dispatching method, characterized in that: The following steps are involved: Obtain user-side load demand and electricity price information for the current time period; Preprocessing the load demand and electricity price information, predicting the load demand in a future time period through an LSTM network according to the load demand, and obtaining a predicted load demand; predicting future electricity price changes through time series analysis according to the electricity price information, and obtaining a predicted electricity price; An optimization goal is set, and an energy allocation scheme is obtained according to the user-side load demand and the electricity price information of the current time period by using a particle swarm optimization algorithm, wherein the objective function of the particle swarm optimization algorithm includes penalties for energy consumption and power balance, and a scheduling scheme is obtained according to the predicted load demand and predicted electricity price by using an MPC algorithm, wherein the objective function of the MPC algorithm includes penalties for electricity price and voltage deviation; The power output of the user-side device is adjusted in real time according to the energy allocation scheme, and the voltage value of the user-side is adjusted in real time according to the scheduling scheme to complete the power scheduling.

2. The intelligent power dispatching method according to claim 1, characterized in that: The optimization objectives include economic objectives, power balance objectives and operational reliability objectives. In the process of obtaining an energy allocation plan based on the user-side load demand and the electricity price information of the current time period through a particle swarm optimization algorithm, the power output of the user-side equipment is adjusted in real time according to the economic objective, and the power balance objective and the operational reliability objective are used as constraints; in the process of obtaining a scheduling plan based on the predicted load demand and the predicted electricity price through an MPC algorithm, the voltage value on the user side is adjusted in real time according to the economic objective, and the power balance objective and the operational reliability objective are used as constraints.

3. The intelligent power dispatching method according to claim 2, characterized in that: The operational reliability objectives are: Where V t is the voltage value on the user side during period t, I t is the current value on the user side during period t, P out (t) is the output power of the user side in the period t, V min is the safe lower limit of voltage, V max is the safety upper limit of the voltage value, I min is the safe lower limit of the current value, I max is the safety upper limit of the current value, P max is the maximum output power.

4. The intelligent power dispatching method according to claim 2, characterized in that: The objective function of the economic objective is: In the formula, C t is the electricity price in period t, P t is the power value on the user side within the t period. The power value on the user side includes output power, load demand power and energy storage power. T is the total time range for optimization.

5. The intelligent power dispatching method according to claim 4, characterized in that: The power balance target is: Where P out (t) is the output power of the user side in the period t, P load (t) is the load demand power in the period t, P storage (t) is the energy storage power in the period t, δ t is the power loss during period t.

6. The intelligent electric energy dispatching method according to claim 5, characterized in that: The output power is: Where P max is the maximum output power, C arg is the average electricity price, C t is the electricity price in period t, α and β are adjustment coefficients; The calculation formula of the energy storage power is: P storage (t)=γ·(C max -C t ), Where P storage (t) is the energy storage power in period t, C max is the maximum electricity price during the operating period, and γ is the energy storage regulation coefficient.

7. The intelligent electric energy dispatching method according to claim 1, characterized in that: The future load demand is predicted by the LSTM network model, and the prediction equation of the load demand is: In the formula, is the load demand forecast value at the future time t+Δt, f LSTM is the mapping of the LSTM network model, P load (t), P load (t-1), ..., P load (tn) is the load demand data at different times in the past.

8. The intelligent electric energy dispatching method according to claim 1, characterized in that: The objective function of the particle swarm optimization algorithm is: In the formula, c i is the unit energy consumption output of the i-th device on the user side, P i is the power output of the ith device on the user side, P total is the total power output at the user side, P required is the power required by the user side, and λ1 is the penalty coefficient for power balancing.

9. The intelligent electric energy dispatching method according to claim 1, characterized in that: The objective function of the MPC algorithm is: Where u is the control input on the user side, N is the prediction duration, λ2 is the penalty factor for voltage deviation, V set is the set voltage value, C t is the electricity price in period t, P t is the power value at the user side during the period t, V t is the voltage value at the user side during period t.

10. An intelligent electric energy dispatching system, characterized in that: include: A data acquisition module is used to obtain user-side load demand and electricity price information for the current time period; A data preprocessing module is used to preprocess the load demand and electricity price information, predict the load demand in the future time period through the LSTM network according to the load demand, and obtain the predicted load demand; predict the future electricity price changes through time series analysis according to the electricity price information, and obtain the predicted electricity price; An optimization strategy generation module, which is used to set optimization goals, obtain the optimal energy allocation plan according to the user-side load demand and the electricity price information of the current time period through a particle swarm optimization algorithm, wherein the objective function of the particle swarm optimization algorithm includes penalties for energy consumption and power balance, and obtain the optimal scheduling plan according to the predicted load demand and predicted electricity price through an MPC algorithm, wherein the objective function of the MPC algorithm includes penalties for electricity price and voltage deviation; The scheduling execution module is used to adjust the power output of the user-side equipment in real time according to the energy allocation plan, and adjust the voltage value of the user-side in real time according to the scheduling plan to complete the power scheduling.

Images