Dynamic power management system based on load characteristics of central air conditioner

Abstract

The invention relates to the technical field of central air-conditioning system management, and discloses a dynamic power management system based on central air-conditioning load characteristics, which mainly comprises a load characteristic analysis module, a power demand prediction module, an energy distribution optimization module and a control execution module. The load characteristic analysis module classifies load characteristics by analyzing historical operation data of the central air-conditioning system; the power demand prediction module predicts future power demand in combination with the load characteristic classification result and weather forecast data; the energy distribution optimization module adopts a genetic algorithm to generate an optimal energy distribution scheme according to the predicted power demand and the current power supply condition; and the control execution module adjusts the operation of the central air-conditioning system according to the optimization result. According to the system, by accurately analyzing the load characteristics, predicting the power demand and optimizing the energy distribution, the power cost is effectively reduced, the power demand of the central air-conditioning system is met, the energy utilization efficiency is improved, and remarkable economic and environment-friendly benefits are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of central air-conditioning system management, and in particular to a dynamic power management system based on central air-conditioning load characteristics. Background Art

[0002] With the continuous expansion of modern building scale and the increasing complexity of functions, the energy consumption problem of central air-conditioning system, as an important part of building energy consumption, has become increasingly prominent. During the operation of central air-conditioning system, its load characteristics are affected by many factors, such as indoor and outdoor temperature, humidity, temperature difference changes, human flow and the operation status of the equipment itself. These factors cause the load demand of central air-conditioning system to show dynamic characteristics, and traditional static power management methods can no longer meet the needs of high efficiency and energy saving.

[0003] In traditional central air-conditioning power management, fixed operation strategies and energy allocation schemes are often used, which cannot be dynamically adjusted according to actual load changes. This management method may lead to energy waste when the load is low, and may not meet the power demand of the system when the load is high, thus affecting the normal operation of the central air-conditioning system and the comfort level in the building.

[0004] In order to solve this problem, people began to study dynamic power management systems based on load characteristics. Most of the existing systems simply adjust according to the current load status, lacking the prediction of future load changes and the optimization of energy distribution. This results in the system being unable to respond in a timely manner when faced with sudden load changes, and the problem of low energy utilization efficiency still exists. Therefore, there is an urgent need for a power management system that can dynamically manage based on the load characteristics of central air conditioning. The system needs to be able to analyze the load characteristics of the central air conditioning system in real time, predict future power demand, and optimize the energy allocation plan based on the current power supply status to minimize power costs and meet the power demand of the central air conditioning system. Summary of the invention

[0005] The object of the present invention is to provide a dynamic power management system based on the central air-conditioning load characteristics to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a dynamic power management system based on central air-conditioning load characteristics, the system comprising:

[0007] The load characteristic analysis module constructs a load characteristic analysis model. The input data of the load characteristic analysis model is the historical operation data of the central air-conditioning system, including temperature, humidity, indoor and outdoor temperature difference, human flow and equipment operation status. The output data is the load characteristic classification result of the central air-conditioning system, including high load, medium load and low load. The model training uses a central air-conditioning system operation data set containing different load states, and the corresponding label is the load state category.

[0008] The power demand prediction module constructs a power demand prediction model, the input data of which are the output results of the load characteristic analysis module and the weather forecast data, and the output data are the predicted power demand of the central air-conditioning system in the future period; the model training uses a time series data set containing load characteristic classification results and weather forecast data, and the corresponding label is the actual power consumption of the central air-conditioning system at each time point;

[0009] The energy distribution optimization module uses a genetic algorithm to build an optimization model. The input data of the optimization model are the prediction results of the power demand prediction model and the current power supply status. The output data is the optimal energy distribution plan to minimize the power cost and meet the power demand of the central air-conditioning system. The model training uses a data set containing the predicted power demand and the actual power supply status, and the corresponding labels are the cost and efficiency indicators of energy distribution.

[0010] The control execution module is connected to the energy distribution optimization module and adjusts the operating parameters and energy distribution of the central air-conditioning system according to the optimization results.

[0011] Preferably, the load characteristic analysis model is constructed using a support vector machine (SVM) algorithm, and its network structure includes:

[0012] Input layer: receiving historical operation data of the central air-conditioning system;

[0013] Feature mapping layer: maps the input data to a high-dimensional space through the RBF kernel function;

[0014] Support vector layer: finds the optimal hyperplane in high-dimensional space to classify load characteristics;

[0015] Output layer: Output the load characteristics classification results of the central air-conditioning system.

[0016] Preferably, the training step of the load characteristic analysis model includes:

[0017] S101: Initialize the parameters of the SVM model, including kernel function parameters, penalty parameters and learning rate;

[0018] S102: Divide the data set used for training into a training set and a validation set;

[0019] S103: Iteratively train the load characteristic analysis model using the training set. In each iteration, the model output is calculated and the loss is calculated according to the loss function. The loss function adopts the Hinge loss function, and the calculation formula is:

[0020]

[0021] Among them, y i is the true category label, f(x i ) is the output of the model decision function, and N is the number of samples;

[0022] S104: Update the model parameters by gradient descent method according to the loss value;

[0023] S105: Repeat steps S103 to S104 until the loss converges or a preset number of iterations is reached;

[0024] S106: Use the validation set to validate the trained model and evaluate the classification accuracy of the model;

[0025] S107: When the model performance reaches the preset standard, the model parameters are saved to obtain a trained load characteristic analysis model.

[0026] Preferably, the power demand forecasting model is constructed using a feedforward neural network FNN algorithm, and its network structure includes:

[0027] Input layer: receives load characteristic classification results and weather forecast data;

[0028] Hidden layer: contains multiple layers of neurons, each layer of neurons is followed by a ReLU activation function to extract features;

[0029] Output layer: Output the predicted power demand value of the central air-conditioning system in the future.

[0030] Preferably, the training step of the power demand prediction model includes:

[0031] S201: Standardize the input data;

[0032] S202: Initialize the parameters of the FNN model, including weights, bias terms, and learning rates;

[0033] S203: Divide the data set into a training set and a validation set;

[0034] S204: Iteratively train the model using the training set, and in each iteration, calculate the model output and calculate the loss according to the loss function;

[0035] S205: Update the model parameters through the back propagation algorithm and the gradient descent method according to the loss value;

[0036] S206: Repeat steps S204 to S205 until the loss converges or a preset number of iterations is reached;

[0037] S207: Use the validation set to validate the trained model and evaluate the prediction accuracy of the model;

[0038] S208: When the model performance reaches the preset standard, the model parameters are saved to obtain a trained power demand forecasting model.

[0039] Preferably, in step S204, a mean square error (MSE) function is used as a loss function, and its calculation formula is:

[0040]

[0041] Among them, y i is the actual power demand of the ith sample, is the predicted power demand of the i-th sample, and M is the number of samples.

[0042] Preferably, the genetic algorithm of the energy distribution optimization model includes:

[0043] Initialization step: Generate an initial population, where each individual represents an energy allocation scheme;

[0044] Fitness evaluation step: According to the power demand forecast results and the current power supply status, the fitness of each individual is calculated to represent the cost and efficiency indicators of energy allocation;

[0045] Selection step: Roulette selection method is used to select individuals for reproduction based on fitness;

[0046] Crossover step: Use the single-point crossover method to generate new individuals;

[0047] Mutation step: Randomly mutate some individuals to increase population diversity;

[0048] Iteration step: Repeat the fitness evaluation, selection, crossover and mutation steps until the preset number of iterations is reached or the optimal solution is found;

[0049] Output step: Output the optimal energy allocation plan.

[0050] Preferably, the calculation formula of the fitness is:

[0051] F=αC+βE

[0052] Among them, C is the total cost of energy allocation, E is the efficiency index of energy allocation, and α and β are weight coefficients used to balance the importance of cost and efficiency.

[0053] Preferably, the time resolution of the historical operation data is not less than once per hour, and the data quality is preprocessed, including missing value filling and outlier detection.

[0054] Preferably, it also includes a data monitoring module for real-time monitoring of the operating data and power supply status of the central air-conditioning system, and inputting the data into the load characteristic analysis module and the power demand prediction module to achieve closed-loop control of dynamic power management.

[0055] Preferably, the training step of the strategy adjustment model includes:

[0056] Step 301: Collect training data, including the difference between the planning results output by the path planning model and the behavior of the logistics robot in the actual task execution process, including path deviation, time delay and energy consumption difference; at the same time, collect the optimal task execution strategy under the given planning result and actual behavior difference annotated by experts as the label of the training data;

[0057] Step 302: construct a random forest RF algorithm model and set RF parameters, including the number of trees, maximum features, and maximum depth;

[0058] Step 303: Use the training data to train the RF algorithm model, and improve the model performance through the bagging method, so that the model can accurately predict the optimal task execution strategy under a given input difference;

[0059] Step 304: After the training is completed, a strategy adjustment model is obtained.

[0060] Preferably, a specific execution strategy of the model output is generated based on the optimal path and task execution strategy output by the path planning model, and relevant logistics knowledge content is created and displayed for the logistics robot, including strategy cases, operating specifications and safety guidelines for different path planning.

[0061] Preferably, the present invention also includes a multi-robot collaboration module, which utilizes a path planning model to plan an optimal path that avoids collisions and enables efficient collaboration based on the real-time status and location information of multiple logistics robots, and further utilizes a task execution strategy generation model to generate a collaborative execution strategy for multiple robots based on the planned path.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] The present invention uses a load characteristic analysis module to accurately classify the load status (high load, medium load, low load) of the central air-conditioning system, providing accurate basic data for subsequent energy management. The power demand forecasting module combines load characteristics and weather forecast data to accurately predict power demand in the future, thereby avoiding energy waste and oversupply. The energy allocation optimization module uses a genetic algorithm to dynamically adjust the energy allocation plan based on the prediction results and the current power supply status, ensuring that while meeting the power demand of the central air-conditioning system, the power cost is minimized and the energy utilization efficiency is significantly improved.

[0064] This system can analyze the load characteristics of the central air-conditioning system in real time, and dynamically adjust the power demand and energy allocation plan according to the load changes and fluctuations in the external environment (such as weather). This dynamic adjustment capability enables the system to adapt to the energy management needs in different scenarios, whether it is commercial buildings, office spaces or public facilities, to achieve efficient energy utilization. Through accurate load characteristic analysis and power demand forecasting, this system can avoid unnecessary energy waste and reduce electricity costs during peak hours. The energy distribution optimization module aims to minimize electricity costs, ensuring that overall operating costs are reduced and economic benefits are improved while meeting demand.

[0065] This system can ensure that the central air-conditioning system can operate stably under various load conditions and provide a comfortable indoor environment. By dynamically adjusting operating parameters and energy distribution, the system can reduce fault downtime and energy shortages, and improve user satisfaction and comfort. The present invention helps to reduce energy consumption and carbon emissions by improving energy utilization efficiency and reducing operating costs, which is in line with the concept of sustainable development. Against the background of global energy shortages and increasing awareness of environmental protection, the promotion and application of this system will play a positive role in promoting green buildings and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a working principle diagram of the dynamic power management system based on central air-conditioning load characteristics described in the present invention;

[0067] Figure 2 Training flow chart for load characteristic analysis model;

[0068] Figure 3 Diagram of the training process of the power demand prediction model. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] See also Figure 1-3 The present invention provides a technical solution: a dynamic power management system based on central air-conditioning load characteristics, the system comprising:

[0071] Implementation of the load characteristic analysis module: Construct a load characteristic analysis model, which takes the historical operation data of the central air-conditioning system as input, including parameters such as temperature, humidity, indoor and outdoor temperature difference, human flow, and equipment operation status. Through data preprocessing and feature extraction, the historical operation data is converted into a format that the model can recognize. The model is trained using a data set containing the operation of the central air-conditioning system under different load states, and the corresponding labels are load state categories (high load, medium load, low load). After the model training is completed, the load characteristic classification results of the central air-conditioning system are output based on the real-time input data.

[0072] Implementation of the power demand forecasting module: Build a power demand forecasting model whose input data is the output of the load characteristic analysis module and weather forecast data. Fusion the load characteristic classification results and weather forecast data to form time series data. Use the time series data set containing the load characteristic classification results and weather forecast data to train the model, and the corresponding label is the actual power consumption of the central air-conditioning system at each time point. After the model training is completed, the power demand of the central air-conditioning system in the future period is predicted based on the real-time input data.

[0073] Implementation of the energy distribution optimization module: Genetic algorithm is used to build an optimization model. The input data of the model are the prediction results of the power demand prediction model and the current power supply status. According to the predicted power demand and the current power supply status, the optimization goal is set to minimize the power cost and meet the power demand of the central air-conditioning system. The optimization model is trained using a data set containing predicted power demand and actual power supply status, and the corresponding labels are the cost and efficiency indicators of energy distribution. After the model training is completed, the optimal energy distribution plan is output based on the real-time input data.

[0074] Implementation of the control execution module: The control execution module is connected to the energy distribution optimization module and receives the optimization results as input. According to the optimization results, the operating parameters and energy distribution of the central air-conditioning system are adjusted to achieve optimal energy utilization and power cost control.

[0075] The present invention will be further described below in conjunction with Examples 1 to 3:

[0076] Embodiment 1:

[0077] The load characteristic analysis model is constructed using the support vector machine (SVM) algorithm, and its detailed implementation is as follows:

[0078] 1.1 Network structure:

[0079] Input layer: Receives historical operating data of the central air-conditioning system, including temperature, humidity, indoor and outdoor temperature difference, passenger flow, and equipment operating status.

[0080] Feature mapping layer: The input data is mapped to a high-dimensional space through a radial basis function (RBF) kernel function in order to find linearly separable features in the high-dimensional space.

[0081] Support vector layer: finds the optimal hyperplane in the high-dimensional space that maximizes the interval between different load characteristic categories and is used to classify load characteristics.

[0082] Output layer: Output the load characteristics classification results of the central air-conditioning system, including high load, medium load and low load.

[0083] 1.2 Training steps:

[0084] S101: Initialize the parameters of the SVM model. This includes the parameters of the RBF kernel function (such as the γ value), the penalty parameter C, and the learning rate (if online learning or incremental learning is used). The selection of these parameters will affect the performance and convergence speed of the model.

[0085] S102: Divide the training data set into a training set and a validation set. The training set is used to train the model, and the validation set is used to evaluate the performance of the model to ensure that the model can perform well on unseen data.

[0086] S103: Iteratively train the load characteristic analysis model using the training set. In each iteration, the output of the model is calculated, and the loss is calculated according to the Hinge loss function. The calculation formula of the Hinge loss function is:

[0087]

[0088] Among them, y i is the true category label, f(x i ) is the output of the model decision function, N is the number of samples, and the loss function encourages the model to give higher confidence to correctly classified samples while imposing penalties on incorrectly classified samples.

[0089] S104: Update the model parameters by gradient descent or other optimization algorithms according to the loss value. This step aims to minimize the loss function so that the model can classify the load characteristics more accurately.

[0090] S105: Repeat steps S103 to S104 until the loss converges or the preset number of iterations is reached. In this process, the performance of the model will gradually improve and the loss value will gradually decrease.

[0091] S106: Use the validation set to validate the trained model. Evaluate the classification accuracy of the model, that is, the proportion of correctly classified samples to the total number of samples. This step is used to verify the generalization ability of the model.

[0092] S107: When the model performance reaches a preset standard (such as the classification accuracy exceeds a certain threshold), the model parameters are saved to obtain a trained load characteristic analysis model, which can be used in actual load characteristic classification tasks.

[0093] Embodiment 2:

[0094] The present invention adopts a feedforward neural network (FNN) algorithm to construct a power demand forecasting model, and its detailed implementation is as follows:

[0095] 2.1 Network structure:

[0096] Input layer: Receives load characteristic classification results (such as high load, medium load, low load) and weather forecast data (such as temperature, humidity, wind speed, etc.). These data are used as input features of the model to predict future power demand.

[0097] Hidden layer: contains multiple layers of neurons, each layer of neurons is followed by a ReLU (Rectified Linear Unit) activation function. The ReLU function can introduce nonlinearity, enabling the model to learn complex feature representations. The design of multiple hidden layers can extract high-level features from the input data and improve the model's predictive ability.

[0098] Output layer: Output the power demand forecast value of the central air-conditioning system in the future. This forecast value is a continuous variable, indicating the power that the central air-conditioning system may consume in a certain period of time in the future.

[0099] 2.2 Training steps:

[0100] S201: Standardize the input data. Since the dimensions and ranges of the input data may be different, standardization can convert the data to the same scale, thereby improving the training efficiency and prediction accuracy of the model.

[0101] S202: Initialize the parameters of the FNN model. This includes weights, bias terms, and learning rates. Weights and bias terms are learnable parameters in the model, and the learning rate controls the step size of the model parameter update.

[0102] S203: Divide the data set into a training set and a validation set. The training set is used to train the model, and the validation set is used to evaluate the performance of the model to ensure that the model can perform well on unseen data.

[0103] S204: Iteratively train the model using the training set. In each iteration, the output of the model is calculated, and the loss is calculated according to the mean square error (MSE) function. The calculation formula of the MSE function is:

[0104]

[0105] Among them, y i is the actual power demand of the ith sample, is the predicted power demand of the i-th sample, and M is the number of samples. The MSE function measures the average squared difference between the predicted value and the true value, and is a commonly used loss function in regression problems.

[0106] S205: Update the model parameters according to the loss value through the back propagation algorithm and the gradient descent method. The back propagation algorithm calculates the gradient of the loss function with respect to the model parameters, and the gradient descent method updates the parameters in the opposite direction of the gradient to minimize the loss function.

[0107] S206: Repeat steps S204 to S205 until the loss converges or the preset number of iterations is reached. In this process, the performance of the model will gradually improve and the loss value will gradually decrease.

[0108] S207: Use the validation set to validate the trained model. Evaluate the prediction accuracy of the model, such as using indicators such as mean square error and mean absolute error. This step is used to verify the generalization ability of the model and ensure that the model can accurately predict power demand in practical applications.

[0109] S208: When the model performance reaches the preset standard (such as the prediction accuracy is lower than a certain threshold), the model parameters are saved to obtain a trained power demand prediction model. This model can be used in actual power demand prediction tasks to provide accurate data support for energy allocation optimization.

[0110] Embodiment 3:

[0111] The present invention adopts genetic algorithm to construct energy distribution optimization model, and the specific steps of construction include:

[0112] Initialization step: Generate an initial population, where each individual in the population represents an energy allocation scheme. These schemes may include power allocation for central air-conditioning systems in different time periods, strategies for enabling and disabling backup power supplies, etc. The size of the initial population can be determined based on the complexity of the problem and computing resources to ensure sufficient search space and population diversity.

[0113] Fitness evaluation steps: Calculate the fitness of each individual based on the power demand forecast results and the current power supply status. The fitness function comprehensively considers the total cost (C) and efficiency index (E) of energy distribution, and its calculation formula is:

[0114] F=αC+βE

[0115] Among them, C is the total cost of energy allocation, which may include electricity purchase cost, equipment operation and maintenance cost, etc.; E is the efficiency index of energy allocation, which may represent energy utilization rate, system stability, etc.; α and β are weight coefficients used to balance the importance of cost and efficiency in fitness evaluation. By calculating fitness, the advantages and disadvantages of each energy allocation scheme can be quantified, providing a basis for subsequent selection, crossover and mutation steps.

[0116] Selection step: Use the roulette wheel selection method to select individuals for reproduction based on their fitness. The roulette wheel selection method allocates selection probabilities based on the fitness ratio of the individuals. Individuals with higher fitness have a greater probability of being selected, and are more likely to pass on their good genes to the next generation.

[0117] Crossover step: Use the single-point crossover method to generate new individuals. The single-point crossover method randomly selects a crossover point and exchanges the gene fragments of the two parent individuals after this point to generate two new offspring individuals. Through the crossover operation, the excellent genes of the parent individuals can be combined to generate offspring individuals with potentially better performance.

[0118] Mutation step: Randomly mutate some individuals to increase population diversity. The mutation operation can randomly change the values ​​of certain genes in individuals, or use other mutation strategies to introduce new gene combinations. Through the mutation operation, new areas in the search space can be explored to avoid the algorithm falling into the local optimal solution.

[0119] Iteration step: Repeat the fitness evaluation, selection, crossover and mutation steps until the preset number of iterations is reached or the optimal solution is found. In each iteration, a new population is generated, and fitness evaluation is used to select excellent individuals for reproduction and mutation. Through continuous iteration, the energy allocation scheme can be gradually optimized to improve the cost and efficiency performance of the system.

[0120] Output step: Output the optimal energy allocation plan, which has the highest fitness value, that is, it maximizes efficiency while minimizing costs. The optimal plan can be used as a decision-making basis for actual energy allocation and guide the power dispatching and energy management of the central air-conditioning system.

[0121] In order to ensure the accuracy of power demand forecasting and energy allocation optimization, the time resolution of historical operation data is not less than once per hour, which means that the system can obtain the operating status data of the central air-conditioning system every hour, such as temperature, humidity, energy consumption, etc., to provide a detailed data basis for subsequent load characteristic analysis and power demand forecasting. In the historical operation data, some data may be missing due to various reasons (such as equipment failure, data transmission error, etc.). The system uses appropriate methods (such as interpolation, mean filling method or prediction filling method based on machine learning) to fill the missing values ​​to ensure the integrity and continuity of the data. The present invention integrates an outlier detection mechanism to identify and process outliers in the data through statistical methods (such as the 3σ rule) or machine learning algorithms (such as isolation forests, support vector machines, etc.). Outliers may be caused by factors such as equipment failure, operating errors or external interference, and processing them can improve the quality and reliability of the data.

[0122] In order to achieve closed-loop control of dynamic power management, the present invention also includes a data monitoring module, whose specific functions are as follows:

[0123] Real-time monitoring: The data monitoring module can collect the operating data and power supply status of the central air-conditioning system in real time. These data include but are not limited to the system's energy consumption, temperature, humidity, equipment status, and the voltage and current of the power network.

[0124] Data transmission and processing: The collected real-time data is transmitted to the system's data processing center through the communication network. In the data processing center, the data is cleaned, verified, and formatted before being input into the load characteristics analysis module and the power demand forecasting module.

[0125] Closed-loop control implementation:

[0126] Load characteristics analysis: The load characteristics analysis module uses real-time data and historical data to conduct in-depth analysis of the load characteristics of the central air-conditioning system. Through analysis, the system's load pattern, peak period, and the relationship between load and external environmental factors (such as weather, time, etc.) can be identified.

[0127] Power demand forecast: The power demand forecast module uses advanced algorithms (such as feedforward neural network FNN, etc.) to forecast future power demand based on the results of load characteristic analysis and real-time data. The forecast results can provide decision support for energy distribution optimization.

[0128] Energy distribution optimization: Based on the power demand forecast results and the current power supply situation, the energy distribution optimization module uses optimization methods such as genetic algorithms to optimize and adjust the energy distribution plan of the central air-conditioning system. The optimized plan is executed through the control system to achieve closed-loop control of dynamic power management.

[0129] Through the above implementation, the present invention can ensure the quality and integrity of historical operation data, and provide a reliable data basis for load characteristic analysis and power demand forecasting. At the same time, through the real-time monitoring and closed-loop control mechanism of the data monitoring module, the system can dynamically adjust the energy allocation plan of the central air-conditioning system to improve the energy efficiency and economy of the system.

[0130] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0131] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic power management system based on central air conditioning load characteristics, characterized in that: The system comprises: The load characteristic analysis module constructs a load characteristic analysis model. The input data of the load characteristic analysis model is the historical operation data of the central air-conditioning system, including temperature, humidity, indoor and outdoor temperature difference, human flow and equipment operation status. The output data is the load characteristic classification result of the central air-conditioning system, including high load, medium load and low load. The model training uses a central air-conditioning system operation data set containing different load states, and the corresponding label is the load state category. The power demand prediction module constructs a power demand prediction model, the input data of which are the output results of the load characteristic analysis module and the weather forecast data, and the output data are the predicted power demand of the central air-conditioning system in the future period; the model training uses a time series data set containing load characteristic classification results and weather forecast data, and the corresponding label is the actual power consumption of the central air-conditioning system at each time point; The energy distribution optimization module uses a genetic algorithm to build an optimization model. The input data of the optimization model are the prediction results of the power demand prediction model and the current power supply status. The output data is the optimal energy distribution plan to minimize the power cost and meet the power demand of the central air-conditioning system. The model training uses a data set containing the predicted power demand and the actual power supply status, and the corresponding labels are the cost and efficiency indicators of energy distribution. The control execution module is connected to the energy distribution optimization module and adjusts the operating parameters and energy distribution of the central air-conditioning system according to the optimization results.

2. The dynamic power management system based on central air conditioning load characteristics according to claim 1 is characterized in that: The load characteristic analysis model is constructed using the support vector machine (SVM) algorithm, and its network structure includes: Input layer: receiving historical operation data of the central air-conditioning system; Feature mapping layer: maps the input data to a high-dimensional space through the RBF kernel function; Support vector layer: finds the optimal hyperplane in high-dimensional space to classify load characteristics; Output layer: Output the load characteristics classification results of the central air-conditioning system.

3. The dynamic power management system based on central air conditioning load characteristics according to claim 2 is characterized in that: The training steps of the load characteristic analysis model include: S101: Initialize the parameters of the SVM model, including kernel function parameters, penalty parameters and learning rate; S102: Divide the data set used for training into a training set and a validation set; S103: Iteratively train the load characteristic analysis model using the training set. In each iteration, the model output is calculated and the loss is calculated according to the loss function. The loss function adopts the Hinge loss function, and the calculation formula is: Among them, y i is the true category label, f(x i ) is the output of the model decision function, and N is the number of samples; S104: Update the model parameters by gradient descent method according to the loss value; S105: Repeat steps S103 to S104 until the loss converges or a preset number of iterations is reached; S106: Use the validation set to validate the trained model and evaluate the classification accuracy of the model; S107: When the model performance reaches the preset standard, the model parameters are saved to obtain a trained load characteristic analysis model.

4. The dynamic power management system based on central air conditioning load characteristics according to claim 1 is characterized in that: The power demand forecasting model is constructed using a feedforward neural network FNN algorithm, and its network structure includes: Input layer: receives load characteristic classification results and weather forecast data; Hidden layer: contains multiple layers of neurons, each layer of neurons is followed by a ReLU activation function to extract features; Output layer: Output the predicted power demand value of the central air-conditioning system in the future.

5. The dynamic power management system based on central air conditioning load characteristics according to claim 4 is characterized in that: The training steps of the power demand prediction model include: S201: Standardize the input data; S202: Initialize the parameters of the FNN model, including weights, bias terms, and learning rates; S203: Divide the data set into a training set and a validation set; S204: Iteratively train the model using the training set, and in each iteration, calculate the model output and calculate the loss according to the loss function; S205: Update the model parameters through the back propagation algorithm and the gradient descent method according to the loss value; S206: Repeat steps S204 to S205 until the loss converges or a preset number of iterations is reached; S207: Use the validation set to validate the trained model and evaluate the prediction accuracy of the model; S208: When the model performance reaches the preset standard, the model parameters are saved to obtain a trained power demand forecasting model.

6. The dynamic power management system based on central air conditioning load characteristics according to claim 5 is characterized in that: In step S204, the mean square error MSE function is used as the loss function, and its calculation formula is: Among them, y i is the actual power demand of the i-th sample, is the predicted power demand of the i-th sample, and M is the number of samples.

7. The dynamic power management system based on central air conditioning load characteristics according to claim 1 is characterized in that: The genetic algorithm of the energy distribution optimization model includes: Initialization step: Generate an initial population, where each individual represents an energy allocation scheme; Fitness evaluation step: According to the power demand forecast results and the current power supply status, the fitness of each individual is calculated to represent the cost and efficiency indicators of energy allocation; Selection step: Roulette selection method is used to select individuals for reproduction based on fitness; Crossover step: Use the single-point crossover method to generate new individuals; Mutation step: Randomly mutate some individuals to increase population diversity; Iteration step: Repeat the fitness evaluation, selection, crossover and mutation steps until the preset number of iterations is reached or the optimal solution is found; Output step: Output the optimal energy allocation plan.

8. The dynamic power management system based on central air conditioning load characteristics according to claim 7 is characterized in that: The calculation formula of the fitness is: F=αC+βE Among them, C is the total cost of energy allocation, E is the efficiency index of energy allocation, and α and β are weight coefficients used to balance the importance of cost and efficiency.

9. The dynamic power management system based on central air conditioning load characteristics according to claim 1 is characterized in that: The time resolution of the historical operation data is not less than once per hour, and the data quality has been preprocessed, including missing value filling and outlier detection.

10. The dynamic power management system based on central air conditioning load characteristics according to claim 1, characterized in that: It also includes a data monitoring module for real-time monitoring of the operating data and power supply status of the central air-conditioning system, and inputting the data into the load characteristic analysis module and the power demand forecasting module to achieve closed-loop control of dynamic power management.

Images