Load identification method of convolutional neural network based on improved VI trajectory

Abstract

The invention discloses a convolutional neural network load identification method based on an improved VI trajectory, and relates to the technical field of power load identification. On the basis of obtaining an independent load current waveform, a voltage waveform is combined, a traditional single-cycle VI track is improved into a mode of drawing a multi-cycle VI track, meanwhile, instantaneous power, a current change rate and average value characteristics of multi-cycle current effective values are introduced, dual color coding improvement of the VI track and the background of the VI track is achieved, and the multi-cycle VI track is obtained. Therefore, the load characteristic characterization capability contained in the VI track is improved. And taking the improved VI track as the input of a VGGNet network framework in the CNN to realize load identification. Aiming at the influence of VGGNet network hyper-parameters on the identification precision, a particle swarm optimization (PSO) algorithm is selected to optimize the hyper-parameters in a network training process so as to improve the load identification accuracy of the model.

Description

Technical Field

[0001] The invention relates to the technical field of electric load identification, in particular to a load identification method based on a convolutional neural network with an improved VI trajectory. Background Art

[0002] With the development of smart grids, the traditional power industry has shifted towards a highly intensive, knowledge-based and technical direction. The power grid has become one of the largest and most complex machines in the world. At the same time, non-intrusive load monitoring has received widespread attention from industry and academia, and has become a research hotspot for the application of artificial intelligence technology in related industries such as construction and electricity. The effective application of non-intrusive load monitoring technology can significantly reduce energy consumption and improve energy efficiency, but there is a problem of low load recognition rate. Summary of the invention

[0003] The object of the present invention is to provide a load identification method based on a convolutional neural network with an improved VI trajectory, so as to solve the problem of low load identification rate.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a load identification method based on a convolutional neural network with an improved VI trajectory, comprising the following steps:

[0005] S1. Improve VI trajectory

[0006] S1.1. Select 25 consecutive cycles of instantaneous voltage and current signals to draw the VI trajectory;

[0007] S1.2, color-code the VI trajectory: use red to represent the multi-cycle VI trajectory, i.e., the R channel; use green to represent the slope of the straight line segment between adjacent acquisition points of the VI trajectory, i.e., the G channel; use blue to represent the instantaneous power value, i.e., the B channel, thereby generating a color-coded VI trajectory image;

[0008] S1.3. Design color mapping based on the range of voltage and current values. Gradient colors can be used to represent changes in voltage or current.

[0009] S2. Constructing a load identification model based on PSO-CNN

[0010] S2.1 CNN-based load identification model

[0011] The typical network model of CNN - VGGNet network is selected as the load identification model. The VGGNet network contains 16 convolutional layers, 5 pooling layers and 3 fully connected layers. The ReLU function is used as the activation function of the convolutional neural network; the VI trajectory feature map is sent to the input layer, that is, the first convolutional layer. The convolutional layer extracts local features and sends them to the pooling layer for feature compression. The fully connected layer integrates the features. The number of units in the last fully connected layer is set according to the number of load types, and the corresponding load category is output; S2.2 Particle Swarm Optimization PSO process:

[0012] S2.2.1. For the VGG network, select an optimization goal that improves the accuracy of the validation set.

[0013] S2.2.2. Select the number of convolution kernels, step size, pooling kernel size and step size as the parameters to be optimized;

[0014] S2.2.3. Initialize the particle swarm, calculate the fitness of each particle, i.e., the hyperparameter combination; update the position and velocity of the particle according to the fitness, and update the position and velocity of each particle using the following formula:

[0015] v i+1 =v i +c1r1(pbest i -x i )+c2r2(gbest i -x i ) (1)

[0016] x i+1 =x i +v i (2)

[0017] Where i = 1, 2, ..., N, N is the total number of particles in this group; v i is the velocity of the particle; v i The maximum value of v max >0, if v i >v max , then v i =v max ; r1, r2 are random numbers between (0, 1); x i is the current position of the particle; c1 and c2 are learning factors, usually c1=c2=2; pbest i is the best known position of particle i; gbest i is the global optimal position;

[0018] For each particle position x i+1, corresponds to a specific hyperparameter combination, uses this combination to train the VGG network, and updates the particle's pbest according to the evaluation index, while updating the global best position gbest;

[0019] S2.3. Particle swarm optimization of CNN hyperparameters:

[0020] First, the fitness function and the objective function are determined. In each iteration cycle, a particle is selected and the fitness function is used to evaluate and optimize it in detail. The evaluation point with the greatest potential identified in each iteration is added to the historical data set. This process continues until the preset termination condition is met. The parameter set X to be optimized, the objective function f, and the fitness function S are used as the input of the particle swarm optimization algorithm to output the optimal VGGNet network model parameters.

[0021] The specific content of step S1.1 is:

[0022] Normalized voltage and current, the formula is as follows:

[0023]

[0024] Where max|v| is the maximum instantaneous voltage in a cycle; max|i| is the maximum instantaneous current in multiple cycles; v k and i k Respectively represent the voltage and current values ​​of the kth acquisition point in the sequence, Δv k and Δi k are the normalized voltage and current values ​​of the kth acquisition point respectively.

[0025] A VI trajectory is created using the normalized data and the resulting VI trajectory is used as the R channel of the color-coded VI trajectory.

[0026] Use the arctan function to map the slope of the straight line segment to the range of (0, 1) to complete the creation of the G channel;

[0027]

[0028] Among them, K j is the slope of the jth straight line segment; G j is the G channel depth value of the jth straight line segment; i j Indicates the current value of the current data point; i j+1 Indicates the current value of the next data point; v j Indicates the voltage value of the current data point; v j+1 represents the voltage value of the next data point; max|v| is the absolute value of the maximum voltage extracted from the data plotting the VI trajectory; max|i| is the absolute value of the maximum current extracted from the data plotting the VI trajectory;

[0029] G j Map it to the VI track to get the G channel depth value G corresponding to each grid m,n , and then normalize it to get the value G of the G channel of each grid point m,n ';

[0030] max G=max{G 1,1 ,G 1,2 ,G 1,3 ,…,G n,n} (7)

[0031]

[0032] Using instantaneous power value P j Create the B channel:

[0033] p j =i j ·v j (9)

[0034] The instantaneous power value p of the multi-cycle j Mapped to the VI track, the depth value of each grid is p m,n , after which the instantaneous power of multiple cycles is normalized;

[0035] max p=max{p 1,1 ,p 1,2 ,p 1,3 ,…,p n,n} (10)

[0036]

[0037] Among them, p m,n is the power superposition value of each grid in the grid; maxp is the maximum power of all grids; p m,n ' is the value of the B channel.

[0038] The background color is determined by first calculating the average of the RMS values ​​of the current energy of multiple adjacent cycles; secondly, mapping the current amplitude to a predefined color range and calculating the color value using a linear interpolation method within a specific interval.

[0039] The color coding gradually transitions from gray to blue, matching the color to the current amplitude.

[0040] VGGNet adopts a deep and stacked convolutional layer structure, uses small convolution kernels and pooling layers to improve feature extraction capabilities and position invariance, and applies weight decay and data augmentation techniques to control model complexity and enhance generalization capabilities. It uses images with dual color coding based on background and VI trajectories as model input to complete recognition tasks.

[0041] The convolution layer consists of multiple filters, each of which performs a convolution operation with the input data to generate a series of feature maps, thereby capturing the local features of the input data; the definition of the convolution operation is expressed by Formula 12-Formula 15:

[0042]

[0043]

[0044] Among them, X is the source feature map; K is the convolution kernel; Y is the new feature map; Y=X*K.

[0045] The VGGNet network sets a pooling layer after each convolutional layer.

[0046] The beneficial effects of the present invention are as follows: the present invention proposes a convolutional neural network load identification method based on an improved VI trajectory, firstly, through real-time monitoring of voltage and current signals, VI trajectory data is generated. After preprocessing, the time domain and frequency domain features of these data are extracted, which enhances the network's ability to identify load characteristics. By utilizing the hierarchical structure of the convolutional neural network, the subtle features of load changes can be effectively captured, thereby achieving high-accuracy load identification. The real-time monitoring and feedback of non-invasive load monitoring technology enable users to quickly obtain energy consumption data and optimize electricity consumption behavior; it can accurately identify different types of electrical equipment and their usage status, helping users understand the energy consumption characteristics of each device; it can monitor abnormal energy consumption patterns, detect equipment failures or potential problems in advance, and avoid major losses; after integration with the smart home system, the technology realizes automated control, improves electricity efficiency and convenience; through precise monitoring and analysis, users can better utilize renewable energy such as solar energy and promote the application of green energy; data visualization enhances users' energy-saving awareness and encourages them to take effective energy-saving measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is the active power and reactive power diagram of the electric kettle;

[0048] Figure 2 This is the power change diagram of the inductive load from startup to stable operation;

[0049] Figure 3 Identify the block diagram for the load;

[0050] Figure 4This is the VGGNet network structure diagram;

[0051] Figure 5 It is the convolution process diagram;

[0052] Figure 6 Flowchart of optimizing a VGGNet network for particle swarm. DETAILED DESCRIPTION

[0053] Based on the independent load current waveform, the present invention combines the voltage waveform to improve the traditional single-cycle VI trajectory into a multi-cycle VI trajectory, and introduces the average value characteristics of instantaneous power, current change rate and multi-cycle current effective value to achieve dual color coding improvement of the VI trajectory and its background, so as to enhance the load feature characterization capability contained in the VI trajectory. The improved VI trajectory is used as the input of the VGGNet (Visual Geometry Group Net) network framework in CNN to achieve load recognition. In view of the influence of VGGNet network hyperparameters on recognition accuracy, the particle swarm optimization PSO algorithm is used to optimize the hyperparameters during the network training process to improve the load recognition accuracy of the model.

[0054] The load characteristics are analyzed as follows:

[0055] The accuracy of load identification is closely related to the electrical characteristics of the load during operation. Selecting appropriate characteristics can improve the identification effect. By analyzing the load type and characteristics, a load characteristic "dictionary" can be constructed. These characteristics are usually divided into two categories: steady-state and transient. Steady-state characteristics include electrical properties such as power, current, harmonics, etc. when the equipment is in stable operation; transient characteristics involve electrical changes that occur during the process from startup to stable operation, or from stable operation to shutdown, such as power fluctuations and transition time.

[0056] Active power and reactive power in steady state are common features in load identification problems. The calculation formulas of active power P and reactive power Q are as follows:

[0057] P=VIcos(φ) (1)

[0058] Q=VIsin(φ) (2)

[0059] Where V is voltage, I is current, and φ is phase angle.

[0060] During the start and stop process, pure resistive loads mainly show rapid changes in active power and can quickly reach stable power when the switch is in operation. At the same time, this type of load consumes almost no reactive power, such as Figure 1 shown.

[0061] When an inductive load is started, there is usually a synchronous step change in active power and reactive power, and the reactive power continues to exist after starting. Figure 2 As shown in the figure, the changes in active and reactive power during the inductive load from startup to stable operation indicate that the influence of inductance and capacitance components at startup causes a short delay. As the delay ends, the power experiences a second obvious step increase and eventually reaches a steady state, and the power will show a certain degree of volatility during the steady state.

[0062] When analyzing steady-state power characteristics, special attention should be paid to the following aspects: whether there is reactive power consumption; the amplitude and rate of change of active power and reactive power during the switching process of equipment; the stability of power characteristics and whether there are obvious fluctuations during the stable operation stage of load.

[0063] Different types of equipment have significant differences in current characteristics. Therefore, when analyzing current characteristics, the current amplitude, volatility, range of change, and whether there are regular periodic changes are important bases for identifying and distinguishing different equipment.

[0064] Within the same power frequency cycle, the characteristic curve drawn with instantaneous voltage u(t) as the horizontal coordinate and instantaneous current i(t) as the vertical coordinate is called the VI trajectory. Loads with different working principles have obvious differences in the shape of their VI trajectories. Therefore, the VI trajectory is often used for load identification. However, there are also some cases where the VI trajectories of some loads are different in type but relatively similar. In order to reduce the impact of the similarity of VI trajectories on the accuracy of load identification, the present invention has made improvements in the following three aspects: the energy loss problem of the VI trajectory; the difference in instantaneous power changes between different loads; the difference in current change rate between different loads.

[0065] During the process from startup to stable operation, various types of loads present different transient characteristics, such as power fluctuation and duration, due to differences in working principles and internal components.

[0066] 1. Improve VI Tracks

[0067] Compared with other single features, the VI trajectory covers a variety of operating characteristics by combining the changes in voltage and current during the operation of household loads, so that it has a higher degree of discrimination in load identification and performs better than traditional features. In addition, the VI trajectory can be effectively combined with computer vision and non-invasive load monitoring (NILM) tasks. Other features usually require a combination and difference analysis of multiple dimensions, while the VI trajectory converts the problem into an image recognition task and uses mature image processing technology to simplify the processing of multiple signals. Therefore, the present invention selects the VI trajectory as the key feature of load identification and improves it to enhance the amount of information it contains.

[0068] The traditional VI trajectory analysis method maps the VI curve to a grid and uses a binary value of 0 or 1 to represent the presence or absence of VI trajectory information in each cell, thereby converting the load identification task into an image recognition problem. The limitation of this method is that the binary representation cannot accurately reflect the details of the VI trajectory, resulting in low accuracy in the expression of load information and difficulty in capturing more subtle changes and features. Therefore, the present invention proposes an improved method, which aims to improve the feature expression and recognition accuracy of the VI trajectory through a richer form of information expression, thereby overcoming the shortcomings of the traditional method.

[0069] (1) Traditional VI trajectory analysis methods usually only focus on the voltage and current changes within a single cycle, and cannot fully capture the fluctuation characteristics and change patterns of the load in multiple cycles. To solve this problem, the present invention uses the instantaneous voltage and current signals of 25 consecutive cycles to draw the VI trajectory, aiming to more comprehensively reflect the changes of the load in different cycles.

[0070] Specifically, the voltage and current signals are first standardized, and then the VI trajectory is drawn based on the standardized data to more accurately express the dynamic characteristics of the load.

[0071] The normalization formula is as follows:

[0072]

[0073] In the formula, max|v| is the maximum value of instantaneous voltage in multiple cycles; max|i| is the maximum value of instantaneous current in multiple cycles; v m and i m Respectively represent the voltage and current values ​​of the mth acquisition point in the sequence, Δv m and Δi m are the normalized voltage and current values ​​of the mth acquisition point respectively.

[0074] (2) The process of color coding the VI trajectory is to more intuitively display the dynamic characteristics and change patterns of the load. In this method, different color channels are used to represent different characteristic information: the red channel (R) is used to represent the changes in the multi-cycle VI trajectory, reflecting the overall trend and characteristics of the load in multiple cycles; the green channel (G) is used to represent the slope of the straight line segment between adjacent acquisition points of the VI trajectory, revealing the volatility and rate of the load in the instantaneous change process; the blue channel (B) is used to represent the instantaneous power value, intuitively showing the power level of the load at each moment.

[0075] Specifically, the voltage and current data are first normalized, and then the VI trajectory is drawn using these standardized data. In this process, the generated VI trajectory data is used as the red channel of the image to show the characteristics of multi-cycle changes. Through this color coding method, the VI trajectory image can more clearly convey the details of the load changes in different cycles, instantaneous power fluctuations, and the rate of change, thereby improving the visualization effect and feature expression ability of the data, which helps the accuracy and efficiency of the load identification task.

[0076] Use the arctan function to map the slope of the straight line segment to the range of (0, 1) to complete the creation of the G channel;

[0077]

[0078] Among them, K j is the slope of the jth straight line segment; G j is the G channel depth value of the jth straight line segment; i j Indicates the current value of the current data point; i j+1 Indicates the current value of the next data point; v j Indicates the voltage value of the current data point; v j+1 represents the voltage value of the next data point; max|v| is the absolute value of the maximum voltage extracted from the data plotting the VI trajectory; max|i| is the absolute value of the maximum current extracted from the data plotting the VI trajectory.

[0079] G j Map it to the VI track to get the G channel depth value G corresponding to each grid m,n , and then normalize it to get the value G of the G channel of each grid point m,n ';

[0080] max G=max{G 1,1 ,G 1,2 ,G 1,3 ,…,G n,n} (7)

[0081]

[0082] The following instantaneous power value p is used j Create the B channel:

[0083] p j =i j ·v j (9)

[0084] The multi-cycle instantaneous power value p j Mapped to the VI track, the depth value of each grid is p m,n,Afterwards, the multi-period power is normalized.

[0085] max p=max{p 1,1 ,p 1,2 ,p 1,3 ,…,p n,n} (10)

[0086]

[0087] Among them, p m,n is the power superposition value of each grid in the grid; maxp is the maximum power of all grids; p m,n ' is the value of the B channel.

[0088] (3) Considering that different loads may show similar VI trajectories, but the current amplitudes are significantly different, and the current amplitude is the key characteristic to distinguish load types, we color-code the VI trajectory background according to the load current amplitude to highlight the difference in current amplitude. The specific method is as follows:

[0089] First, the average value of the RMS energy of the current signal in 25 consecutive cycles is calculated. Then, the current amplitude is mapped to a predefined color range, and the color is assigned to different amplitude intervals using a linear interpolation method. The color coding gradually transitions from gray to blue, covering the range from the lowest to the highest current amplitude. In this way, the change of background color can reflect the difference in current amplitude, enhancing the correspondence between the VI trajectory image and the load current characteristics.

[0090] 2. Building a load identification model based on PSO-CNN

[0091] Convolutional neural network (CNN) is used as the load identification model to identify VI trajectories. However, due to the large number of hyperparameters in CNN, the selection of hyperparameters will have a great impact on the load identification results, and manual adjustment of hyperparameters is cumbersome. When optimizing the VGG network, particle swarm optimization (PSO) has high global search capabilities and strong adaptability to complex nonlinear hyperparameter spaces. PSO searches for the global optimal solution through multi-particle parallelism. It is easier to implement and adjust than traditional optimization methods. It is suitable for high-dimensional parameter spaces and multi-objective optimization tasks, and can effectively improve the performance and generalization ability of deep learning models. Therefore, using the PSO algorithm to optimize the hyperparameters in CNN can improve the accuracy of load identification. Specifically:

[0092] 2.1 CNN-based load identification model

[0093] The VI trajectory features are converted into images, and then the VI trajectory images are classified using CNN, which is commonly used in image processing, to identify various load types. In the CNN load identification model, Figure 3 As shown, firstly, the VI trajectory diagrams of different load types are classified and labeled, and then supervised training is performed on the VI trajectories of various loads. Through the iteration and optimization process, the difference between the VI trajectory of each load and the VI trajectories of other loads is identified.

[0094] The CNN model is mainly composed of an input layer, a convolutional layer, a pooling layer, and a fully connected layer. The present invention selects the typical network model of CNN - VGGNet network as the load recognition model. The model structure is as follows Figure 4 As shown in the figure, it contains 16 convolutional layers, 5 pooling layers and 3 fully connected layers. The VGGNet network uses the ReLU function as the activation function of the convolutional neural network. Compared with the Sigmoid function, it has better effects and solves the overfitting problem. In addition, VGGNet uses a deep and stacked convolutional layer structure, uses small convolution kernels and pooling layers to improve feature extraction capabilities and position invariance, and applies weight decay and data enhancement techniques to control model complexity and enhance generalization capabilities. Images with dual color coding based on background and VI tracks are used as model input to complete recognition tasks.

[0095] The convolution layer is one of the core layers of VGGNet. It consists of multiple filters. Each filter performs a convolution operation with the input data to generate a series of feature maps, thereby capturing the local features of the input data. Figure 5 The process of sliding the convolution kernel on the original feature map is shown to extract features and generate a new feature map. According to specific needs, the required feature map data can be obtained by adjusting the parameters of the convolution kernel. The definition of the convolution operation can be expressed by Formula 12-Formula 15. The size and step size of the convolution kernel affect the receptive field of the pixels in the feature map, which in turn affects the feature extraction effect of the model. At the same time, the number of convolution kernels also affects whether the network can analyze the input data from multiple angles. Too many kernels may reduce the recognition ability of the network and increase the computing requirements.

[0096]

[0097] Among them, X is the source feature map; K is the convolution kernel; Y is the new feature map; Y=X*K.

[0098] The VGGNet network sets a pooling layer after each convolution layer. The main function of the pooling layer is to reduce the spatial size of the feature map by downsampling, filter out unnecessary samples, thereby retaining key information, reducing model parameters, preventing overfitting, and improving computational efficiency. Setting the appropriate pooling kernel size and step size can reduce the computational complexity of the data while retaining data information.

[0099] During network training, the randomly set loss probability may affect the recognition performance of the model. The fully connected layer is located at the end of the VGGNet network, and the last fully connected layer passes the output to an m-class softmax classifier to perform multi-classification tasks for m different load categories in the dataset.

[0100] In summary, the load identification process based on CNN can be summarized as follows: the VI trajectory feature map is sent to the input layer (i.e., the first convolution layer), the convolution layer extracts local features and sends them to the pooling layer for feature compression, the fully connected layer integrates the features, and the number of fully connected units in the last layer is set according to the number of load types, and the corresponding load category is output. In the process of load identification, an additional "load not turned on" output state is set, that is, according to the number of load types n, the system sets n+1 output units. If the similarity between the VI trajectory to be identified and the VI trajectory feature library is less than 0.5, the system will output the "load not turned on" state instead of the specific load type.

[0101] 2.2 Particle Swarm Optimization PSO Process:

[0102] First, we need to clearly define the optimization goal. For the VGG network, we choose an optimization goal that improves the accuracy of the validation set.

[0103] Secondly, determine the hyperparameters to be optimized, and select the number of convolution kernels, step size, pooling kernel size, and step size as the parameters to be optimized.

[0104] Then, the particle swarm is initialized, and the fitness of each particle, that is, the hyperparameter combination, is calculated. In deep learning, fitness is usually evaluated by training a neural network and calculating the accuracy on the validation set. The position and velocity of the particle are updated according to the fitness. The PSO algorithm updates the position and velocity of each particle using the following formula:

[0105] v i+1 =v i +c1r1(pbest i -x i )+c2r2(gbest i -x i ) (16)

[0106] x i+1 =x i +v i (17)

[0107] Where i = 1, 2, ..., N, N is the total number of particles in this group; v i is the velocity of the particle; v i The maximum value of v max >0, if v i>v max , then v i =v max ; r1, r2 are random numbers between (0, 1); x i is the current position of the particle; c1 and c2 are learning factors, usually c1=c2=2; pbest i is the best known position of particle i; gbest i is the global optimal position;

[0108] For each particle position x i+1 , corresponding to a specific hyperparameter combination, the combination is used to train the VGG network, and the particle pbest is updated according to the evaluation index, and the global best position gbest is updated at the same time. The iteration of PSO can set the maximum number of iterations or terminate after reaching a satisfactory fitness value. Once the termination condition is met, the hyperparameter combination corresponding to the global best particle is selected as the final optimization result. The optimal hyperparameters are applied to the actual training and use of the VGG network to obtain better performance or efficiency. In this way, PSO can effectively search the hyperparameter space of the VGG network, help optimize the network structure, and improve its performance and generalization ability.

[0109] Through these steps, the PSO algorithm can effectively search and optimize the hyperparameter configuration of the VGG network and improve the performance and generalization ability of the load identification model. The advantage of the PSO algorithm lies in its global search capability, which helps to avoid falling into the local optimal solution and quickly converge to a better solution.

[0110] 2.3 Particle Swarm Optimization CNN Hyperparameter Algorithm:

[0111] Regarding the analysis of the influence of hyperparameters on the load identification effect, the number of convolution kernels, step size, pooling kernel size and step size are selected as the parameters to be optimized.

[0112] In the process of particle swarm optimization of the VGGNet network model to optimize its hyperparameters, such as Figure 6 As shown, the fitness function and the objective function are first determined. In each iteration cycle, a particle is selected and evaluated and optimized in detail using the fitness function. The evaluation point with the greatest potential identified in each iteration will be added to the historical data set, and this process will continue until the preset termination condition is met. The input of the particle swarm optimization algorithm in the present invention includes the parameter set to be optimized X, the objective function f, and the fitness function S, and the output is the optimal VGGNet network model parameters.

Claims

1. A load identification method based on a convolutional neural network with improved VI trajectory, characterized in that: The following steps are involved: S1. Improve VI trajectory S1.

1. Select 25 consecutive cycles of instantaneous voltage and current signals to draw the VI trajectory; S1.2, color-code the VI trajectory: use red to represent the multi-cycle VI trajectory, i.e., the R channel; use green to represent the slope of the straight line segment between adjacent acquisition points of the VI trajectory, i.e., the G channel; use blue to represent the instantaneous power value, i.e., the B channel, thereby generating a color-coded VI trajectory image; S1.

3. Design color mapping based on the range of voltage and current values. Gradient colors can be used to represent changes in voltage or current. S2. Constructing a load identification model based on PSO-CNN S2.1 CNN-based load identification model The typical network model of CNN - VGGNet network is selected as the load recognition model. The VGGNet network contains 16 convolutional layers, 5 pooling layers and 3 fully connected layers. The ReLU function is used as the activation function of the convolutional neural network. The VI trajectory feature map is sent to the input layer, that is, the first convolutional layer. The convolutional layer extracts local features and sends them to the pooling layer for feature compression. The fully connected layer integrates the features. The number of units in the last fully connected layer is set according to the number of load types, and the corresponding load category is output. S2.2 Particle Swarm Optimization PSO process: S2.2.

1. For the VGG network, select an optimization goal that improves the accuracy of the validation set. S2.2.

2. Select the number of convolution kernels, step size, pooling kernel size and step size as the parameters to be optimized; S2.2.3, initialize the particle swarm, and calculate the fitness of each particle, i.e., the hyperparameter combination; Update the position and velocity of the particle according to the fitness, and update the position and velocity of each particle using the following formula: v i+1 =v i +c1 r2 (pbest i -x i )+c2 r2 (gbest i -x i ) (1) x i+1 =x i +v i (2) Where i = 1, 2, ..., N, N is the total number of particles in this group; v i+1 is the updated velocity of the particle; v i is the current velocity of the particle; v i The maximum value of v max >0, if v i >v max , then v i =v max ; r1, r2 are random numbers between (0, 1); x i is the current position of the particle; x i+1 is the updated position of the particle; c1 and c2 are learning factors, usually c1=c2=2; pbest i is the best known position of particle i; gbest i is the global optimal position; For each particle position x i+1 , corresponding to a specific hyperparameter combination, uses this combination to train the VGG network, and updates the particle's pbest according to the evaluation index, while updating the global best position gbest i ; S2.

3. Particle swarm optimization of CNN hyperparameters: First, the fitness function and the objective function are determined. In each iteration cycle, a particle is selected and the fitness function is used to evaluate and optimize it in detail. The evaluation point with the greatest potential identified in each iteration is added to the historical data set. This process continues until the preset termination condition is met. The parameter set X to be optimized, the objective function f, and the fitness function S are used as the input of the particle swarm optimization algorithm to output the optimal VGGNet network model parameters.

2. According to claim 1, a load identification method based on a convolutional neural network with improved VI trajectory is characterized in that: The specific content of step S1.1 is: Normalized voltage and current, the formula is as follows: In the formula, max|v| is the maximum voltage in the data; max|i| is the instantaneous maximum value in the data; v k and i k Respectively represent the voltage and current values ​​of the kth acquisition point in the data, Δv k and Δi k are the normalized voltage and current values ​​of the kth acquisition point respectively.

3. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 2 is characterized in that: A VI trajectory is created using the normalized data and the resulting VI trajectory is used as the R channel of the color-coded VI trajectory.

4. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 3 is characterized in that: Use the arctan function to map the slope of the straight line segment to the range of (0, 1) to complete the creation of the G channel; Among them, K j is the slope of the jth straight line segment; G j is the G channel depth value of the jth straight line segment; i j Indicates the current value of the current data point; i j+1 Indicates the current value of the next data point; v j Indicates the voltage value of the current data point; v j+1 represents the voltage value of the next data point; max|v| is the absolute value of the maximum voltage extracted from the data plotting the VI trajectory; max|i| is the absolute value of the maximum current extracted from the data plotting the VI trajectory; G j Map it to the VI track to get the G channel depth value G corresponding to each grid m,n , and then normalize it to get the value G of the G channel of each grid point m,n '; maxG=max{G 1,1 ,G 1,2 ,G 1,3 ,…,G n,n } (7) 5. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 4 is characterized in that: Using the instantaneous power value p j Create the B channel: p j =i j v j (9) The instantaneous power value p of the multi-cycle j Mapped to the VI track, the depth value of each grid is p m,n , after which the instantaneous power of multiple cycles is normalized; maxp=max{p 1,1 ,p 1,2 ,p 1,3 ,…,p n,n } (10) Among them, p m,n is the power superposition value of each grid in the grid; maxp is the maximum power of all grids; p m,n ' is the value of the B channel.

6. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 5 is characterized in that: The background color is determined by first calculating the average of the RMS value of the current energy of 25 adjacent cycles; secondly, mapping the current amplitude to a predefined color range and calculating the color value using a linear interpolation method within a specific interval.

7. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 6 is characterized in that: The color coding gradually transitions from gray to blue and is divided into 6 intervals: gray-cyan to green, green to orange, orange to yellow, yellow to red, red to purple and purple to blue, achieving a match between color and current amplitude.

8. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 1 is characterized in that: VGGNet adopts a deep and stacked convolutional layer structure, uses small convolution kernels and pooling layers to improve feature extraction capabilities and position invariance, and applies weight decay and data augmentation techniques to control model complexity and enhance generalization capabilities. It uses images with dual color coding based on background and VI trajectories as model input to complete recognition tasks.

9. The load identification method based on a convolutional neural network with improved VI trajectory according to claim 8 is characterized in that: The convolution layer consists of multiple filters, each of which performs a convolution operation with the input data to generate a series of feature maps, thereby capturing the local features of the input data; the definition of the convolution operation is expressed by Formula 12-Formula 15: Among them, X is the source feature map; K is the convolution kernel; Y is the new feature map; Y=X*K.

10. A load identification method based on a convolutional neural network with improved VI trajectory according to claim 9, characterized in that: The VGGNet network sets a pooling layer after each convolutional layer.

Images