Multi-time scale coordination control method for multi-energy power system

Abstract

The invention relates to a multi-time-scale coordination control method for a multi-energy power system. The method comprises the following steps: (a) acquiring and identifying an energy node topology network of the multi-energy power system based on spatial positions of energy nodes in a target area; (b) traversing historical data of the energy node topology network to obtain an energy historical data set of the energy node topology network; (c) performing data preprocessing on the energy historical data set; (d) constructing a Gaussian mixture model by using the preprocessed data set, and initializing a clustering center of the Gaussian mixture model and a prior probability and a parameter of each Gaussian component; (e) outputting an optimal mean value and an optimal variance by using particle swarm optimization; (f) normalizing the probability that the output data in the step (e) comes from different Gaussian distributions; and (g) obtaining scheduling weights of wind energy, solar energy, heat energy and electric energy. And a Gaussian mixture model is constructed by using the preprocessed data set, so that a scheduling weight is output, and electric power coordination control is more accurately carried out.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power dispatching, and particularly relates to a multi-time scale coordinated control method for a multi-energy power system. BACKGROUND

[0002] The multi-energy system refers to a new energy system concept formed by coupling of various energy systems such as cold, heat, electricity and gas in energy production, transmission and use. The multi-energy system makes full use of the mutual aid and complementarity of different forms of energy, improves the economy of the system, enhances the flexibility of the system, increases the reliability of the system and excavates the complementarity of the system. The multi-energy system combines stable power sources such as thermal power and nuclear power to build a stable and reliable energy supply system through multi-energy complementarity and integration of renewable energy such as wind energy and solar energy. The multi-energy complementarity effectively suppresses the intermittency and volatility of new energy through the cooperation of energy storage and regulation, and promotes the consumption of renewable energy.

[0003] At present, the power dispatching of the multi-energy power system has certain limitations in actual application: the dispatching accuracy of various types of energy is poor, and efficient and stable coordinated dispatching is difficult to achieve. SUMMARY

[0004] The application aims to overcome the deficiencies of the prior art and provide a multi-time scale coordinated control method for a multi-energy power system.

[0005] To achieve the above object, the application adopts the technical scheme of a multi-time scale coordinated control method for a multi-energy power system, which comprises:

[0006] (a) obtaining and identifying the energy node topology network of the multi-energy power system based on the spatial positions of the energy nodes in the target area;

[0007] (b) traversing the historical data of the energy node topology network to obtain an energy historical data set of the energy node topology network;

[0008] (c) performing data preprocessing on the energy historical data set;

[0009] (d) constructing a Gaussian mixture model using the preprocessed data set, initializing the cluster centers of the Gaussian mixture model and the prior probability and parameters of each Gaussian component;

[0010] (e) outputting the optimal mean and optimal variance using particle swarm optimization;

[0011] (f) normalizing the probability of the output data from different Gaussian distributions in step (e);

[0012] (g) obtaining the dispatching weights of wind energy, solar energy, heat energy and electric energy; and the multi-energy power system performs power coordinated control according to the dispatching weights.

[0013] Optimally, the data preprocessing of the energy history dataset in step (c) comprises:

[0014] (c1) screening out parameters with high correlation to the multi-energy power system, including output voltage, output current, input power and output power of wind energy, solar energy, thermal energy and electric energy;

[0015] (c2) reconstructing new features of the multi-energy power system usage process, including voltage difference, current difference and scheduling weight ratio.

[0016] Further, step (e) comprises:

[0017] (e1) initializing population parameters using a particle swarm algorithm, each particle representing a set of parameters of the Gaussian component, initializing the initial position and velocity of each particle;

[0018] (e2) calculating the initial fitness value according to the fitness function, obtaining the optimal position p best and the current global optimal position g best of the initial particle;

[0019] (e3) calculating the fitness value of each particle according to the fitness function, comparing the current particle optimal position with the last particle optimal position, and comparing the current global optimal position value with the last current global optimal position;

[0020] (e4) selecting a smaller fitness value to adjust the individual optimal position and global optimal position of the particle; outputting k sets of optimal mean and optimal variance.

[0021] Still further, step (e2) further comprises:

[0022] updating the velocity and position of each particle, the formula being as follows:

[0023] v i (t+1)=ω·v i (t)+c1·r1·(p i -x i )+c2·r2·(g-x i )

[0024] x i (t+1)=x i (t)+v i (t+1)

[0025] wherein v i (t) is the velocity of the i-th particle at the t-th iteration, x i (t) is the position of the i-th particle at the t-th iteration, p iThe history optimal position of the i-th particle, g is the global optimal position, ω is an inertia weight, c1 and c2 are learning factors, and r1 and r2 are random numbers.

[0026] Optimally, in step (b), the energy real-time data set is also acquired.

[0027] Further, in step (d), a preset time scale is also acquired, the energy real-time data set and the preset time scale are input into the Gaussian mixture model, a complementary probability between each energy under the preset time scale is output, a complementary probability set is output, the energy node topology network is identified according to the size of the complementary probability set, and the energy node topology network with complementary path identification is output.

[0028] Due to the use of the above technical solution, the present application has the following advantages compared with the prior art: the multi-energy power system multi-time scale coordinated control method of the present application uses a preprocessed data set to construct a Gaussian mixture model, thereby outputting a scheduling weight, and more accurately performing power coordinated control. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 FIG. 1 is a schematic diagram of the multi-energy power system multi-time scale coordinated control method of the present application. DETAILED DESCRIPTION

[0030] The present application will be further described below in connection with the embodiments shown in the accompanying drawings.

[0031] As shown in the multi-energy power system multi-time scale coordinated control method, it comprises: Figure 1

[0032] (a) acquiring and identifying the energy node topology network of the multi-energy power system based on the spatial positions of the energy nodes in the target region;

[0033] (b) traversing the historical data of the energy node topology network, acquiring an energy historical data set of the energy node topology network, and also acquiring an energy real-time data set;

[0034] (c) data preprocessing the energy historical data set;

[0035] In step (c), the data preprocessing of the energy historical data set comprises:

[0036] (c1) screening out parameters with high correlation to the multi-energy power system, including output voltage, output current, input power and output power of wind energy, solar energy, thermal energy and electric energy;

[0037] (c2) reconstructing new features of the use process of the multi-energy power system, including voltage difference, current difference and scheduling weight ratio.​

[0038] (d) constructing a Gaussian mixture model using the preprocessed dataset, initializing the cluster centers of the Gaussian mixture model and the prior probability, parameters of each Gaussian component;

[0039] A preset timing scale is also obtained, and the energy real-time dataset and the preset timing scale are input into the Gaussian mixture model, and a complementary probability between each energy under the preset timing scale is output, and a complementary probability set is output; the energy node topology network is identified according to the size of the complementary probability set, and an energy node topology network with complementary path identification is output;

[0040] (e) outputting the optimal mean and optimal variance using particle swarm optimization;

[0041] Step (e) includes:

[0042] (e1) initializing population parameters using a particle swarm algorithm, each particle representing a set of parameters of the Gaussian component, and initializing the initial position and speed of each particle;

[0043] (e2) calculating the initial fitness value according to the fitness function, obtaining the optimal position p best and the current global optimal position g best of the initial particle; step (e2) further includes:

[0044] updating the speed and position of each particle, the formula being as follows:

[0045] v i (t+1) = ω·v i (t) + c1·r1·(p i -x i ) + c2·r2·(g-x i )

[0046] x i (t+1) = x i (t) + v i (t+1)

[0047] wherein v i (t) is the speed of the i-th particle at the t-th iteration, x i (t) is the position of the i-th particle at the t-th iteration, p i is the historical optimal position of the i-th particle, g is the global optimal position, ω is the inertia weight, c1 and c2 are learning factors, and r1 and r2 are random numbers;

[0048] (e3) according to the fitness function, the fitness value of each particle is calculated, the current particle optimal position is compared with the last particle optimal position, and the current global optimal position value is compared with the last current global optimal position;

[0049] (e4) selecting a smaller fitness value, thereby adjusting the individual optimal position and the global optimal position of the particle; outputting the optimal mean and the optimal variance of the k groups

[0050] (f) normalizing the probability that the output data in step (e) come from different Gaussian distributions;

[0051] (g) obtaining the scheduling weight of wind energy, solar energy, thermal energy and electric energy; the multi-energy power system performs power coordination control according to the scheduling weight, which is beneficial to improving the precision of power coordination control.

[0052] The above examples are only for illustrating the technical concept and characteristics of the present application, and the purpose is to enable those skilled in the art to understand the content of the present application and implement it, and cannot limit the protection scope of the present application. Any equivalent changes or modifications made according to the spirit and principle of the present application shall be covered within the protection scope of the present application.

Claims

1. A multi-time scale coordinated control method for a multi-energy power system, characterized in that, It comprises: (a) obtaining and identifying the energy node topology network of the multi-energy power system based on the spatial position of the energy nodes in the target area; (b) traversing the historical data of the energy node topology network to obtain an energy historical data set of the energy node topology network; (c) data preprocessing of the energy historical data set; (d) constructing a Gaussian mixture model using the preprocessed data set, initializing the cluster centers of the Gaussian mixture model and the prior probability and parameters of each Gaussian component; (e) outputting the optimal mean and optimal variance using particle swarm optimization; (f) normalizing the probability of the output data from different Gaussian distributions in step (e); (g) obtaining the scheduling weights of wind energy, solar energy, thermal energy, and electric energy; and the multi-energy power system performs power coordination control according to the scheduling weights.

2. The multi-energy power system multi-time scale coordinated control method according to claim 1, characterized in that, In step (c), the data preprocessing of the energy historical data set comprises: (c1) screening out parameters with high correlation to the multi-energy power system, including output voltage, output current, input power, and output power of wind energy, solar energy, thermal energy, and electric energy; (c2) reconstructing new features of the multi-energy power system usage process, including voltage difference, current difference, and scheduling weight ratio.

3. The multi-energy power system multi-time scale coordinated control method according to claim 1 or 2, characterized in that, Step (e) comprises: (e1) initializing population parameters using the particle swarm algorithm, each particle representing a set of parameters of the Gaussian component, and initializing the initial position and velocity of each particle; (e2) calculating an initial fitness value according to the fitness function, obtaining the optimal position p of the initial particle best and the current global optimal position g best ; (e3) calculating the fitness value of each particle according to the fitness function, comparing the current optimal position of the particle with the previous optimal position of the particle, and comparing the current global optimal position value with the previous current global optimal position; (e4) selecting a smaller fitness value to adjust the individual optimal position and global optimal position of the particle; and outputting k sets of optimal mean and optimal variance.

4. The multi-energy power system multi-time scale coordinated control method according to claim 3, characterized in that, Step (e2) further comprises: updating the velocity and position of each particle, as follows: v i (t+1) = ω · v i (t) + c1 · r1(p i - x i ) + c2 · r2 · (g - x i ) x i (t+1) = x i (t) + v i (t+1) where v i (t) is the velocity of the i-th particle at the t-th iteration, x i (t) is the position of the i-th particle at the t-th iteration, p i is the history best position of the i-th particle, g is the global best position, ω is the inertia weight, c1 and c2 are the learning factors, and r1 and r2 are random numbers.

5. The multi-energy power system multi-time scale coordinated control method according to claim 1, characterized in that, In step (b), the energy real-time data set is also obtained.

6. The multi-energy power system multi-time scale coordinated control method according to claim 5, characterized in that, In step (d), a preset time scale is also obtained, and the energy real-time data set and the preset time scale are input into the Gaussian mixture model to output the complementary probability between each energy based on the preset time scale, output a complementary probability set, and identify the complementary path of the energy node topology network according to the size of the complementary probability set, and output the energy node topology network with the complementary path identification.

Images