Data optimization method and system for smart apartment

Abstract

The invention discloses a data optimization method and system for a smart apartment, and the method comprises the steps: S1, collecting multi-source heterogeneous data in the smart apartment, and constructing a hypergraph model; s2, a hypergraph neural network model is constructed, and the hypergraph neural network model is used for extracting high-order features from the multi-source heterogeneous data of the smart apartment and analyzing multivariate relationships among various entities in the apartment; s3, mapping an optimization problem in intelligent apartment management into a quantum Hamiltonian optimization problem; the optimization problem comprises energy distribution optimization; s4, generating a preliminary solution by using a classical optimization algorithm, performing global search by using a quantum stimulated state evolution algorithm to obtain an optimal solution, and performing collaborative optimization; and S5, performing energy scheduling and equipment control on the smart apartment by using an optimal solution obtained by collaborative optimization. According to the method, efficient processing and deep analysis of the multi-source heterogeneous data of the smart apartment are realized, and the accuracy and efficiency of energy management, equipment control and user behavior analysis are remarkably improved.

Description

Technical Field

[0001] The invention relates to a data optimization method for a smart apartment and belongs to the field of intelligent building management systems. Background Art

[0002] With the rapid development of smart cities and the popularization of Internet of Things technology, smart apartments, as an important part of smart cities, have become a hot topic in research on improving their management efficiency and service quality. Traditional apartment management usually relies on preset rules or simple statistical models, which are difficult to effectively play a role in processing multi-source heterogeneous data generated by a large number of smart devices and sensors, resulting in low data utilization and difficulty in meeting the needs of modern residents.

[0003] When processing complex multi-source heterogeneous data, traditional machine learning methods (such as support vector machines, random forests, etc.) and simple deep learning models (such as feedforward neural networks) often fail to fully utilize the high-order relationships and structural information contained in the data. Graph neural networks (GNNs), as a deep learning model that can process graph-structured data, have achieved remarkable results in the fields of social network analysis and knowledge graphs. However, the entities and relationships in smart apartments are more complex, involving multivariate relationships, which are difficult for traditional graph neural networks to fully capture.

[0004] As an extension of GNN, Hypergraph Neural Network (HGN) can process hypergraph structures and capture high-order correlations between multivariate relationships, providing a new solution for optimizing the management data of smart apartments. At the same time, the development of quantum computing technology, especially the quantum excited state evolution algorithm, has made it possible to solve complex combinatorial optimization problems. By mapping the optimization problem of smart apartments into a quantum Hamiltonian, quantum algorithms can be used to solve the global optimal solution.

[0005] In addition, the generative adversarial network (GAN) can improve the generalization ability of the model and adapt to changes in data distribution through adversarial training of the generator and the discriminator. Introducing GAN into the management model of smart apartments can enhance the robustness and adaptability of the model. How to effectively integrate hypergraph neural networks, quantum computing, and adversarial training and apply them to the optimization of management data for smart apartments is a technical problem that needs to be solved. Summary of the invention

[0006] The technical problem to be solved by the present invention is that in the data management of smart apartments, the optimization calculation efficiency of multi-source heterogeneous data processing is low and the real-time performance is poor.

[0007] In order to solve the above technical problems, the present invention provides a data optimization method for a smart apartment, comprising the following steps:

[0008] S1. Collect multi-source heterogeneous data in smart apartments and build a hypergraph model;

[0009] S2. construct a hypergraph neural network model, which is used to extract high-order features from multi-source heterogeneous data of the smart apartment and analyze the multivariate relationships between various entities in the apartment, including rooms, devices, sensors, and users;

[0010] S3. Mapping the optimization problem in smart apartment management into a quantum Hamiltonian optimization problem; the optimization problem includes energy distribution optimization;

[0011] S4. Generate a preliminary solution using a classical optimization algorithm, use a quantum excited state evolution algorithm to perform a global search to obtain an optimal solution, and perform collaborative optimization;

[0012] S5. The optimal solution obtained by collaborative optimization in step S4 Carry out energy scheduling and equipment control for smart apartments.

[0013] The aforementioned data optimization method for a smart apartment includes, in step S1:

[0014] S11. In smart apartments, IoT sensors, smart devices and user terminals are used to collect multi-source heterogeneous data in real time, including environmental parameters, device status, user behavior and energy consumption;

[0015] S12. Preprocessing the collected multi-source heterogeneous data;

[0016] S13. Constructing a hypergraph model of smart apartment based on preprocessed data ,in:

[0017] Node Set : Includes all devices, sensors and users;

[0018] Super edge set : Indicates a multivariate relationship, including multiple devices being used by the same user, or multiple sensors jointly monitoring the same environmental parameter;

[0019] In the aforementioned data optimization method for a smart apartment, in step S2, a multi-layer hypergraph convolutional network is designed in the hypergraph neural network model to learn the high-order feature representation of nodes. The structure and formula of the multi-layer hypergraph convolutional network are as follows:

[0020] For Layer, Node The characteristic expression is:

[0021]

[0022] in: Is a node In the Hidden representation of layers; Contains nodes The hyperedge set of ;

[0023] It is a super edge The set of nodes contained; It is a super edge The weight of and are learnable weight matrices and bias terms; is the activation function, Indicates that node u is in The feature representation of the layer, node u is related to node Through the super edge Connected neighbor nodes, is the node currently being processed, and u is on the same hyperedge Other nodes in the computation node Feature updates.

[0024] In the aforementioned data optimization method for a smart apartment, in step S2, a corresponding loss function is set according to the management task, and the total loss function is:

[0025]

[0026] in: is the number of tasks; It is The loss function of each task; is the task weight; It is against loss; is the weight against loss;

[0027] The management tasks include anomaly detection and device status feedback.

[0028] The aforementioned data optimization method for a smart apartment, in step S2, uses a generator and the discriminator Construct a generative adversarial network with the adversarial loss as:

[0029] L adv (G,D)= E x∼ P data [logD(x)]+ E z∼ P z [log(1-D(G(z)))]

[0030] in, E x ∼ P data [ ⋅ ] Represents the distribution from real data The data sample extracted from , the data samples are the data actually observed in the smart apartment;

[0031] E z ∼ P z [ ⋅ ] Represents the latent variable Distribution The sample drawn from ;

[0032] Represents the discriminator for the sample The judgment result of

[0033] represents the generator from the latent variables Generated samples.

[0034] The aforementioned data optimization method for a smart apartment includes, in step S3: 1) initial state setting: from the initial state To begin with, the initial state is determined by the initial Hamiltonian The ground state is determined, which represents the simplest configuration of the system;

[0035] 2) Construct a target Hamiltonian, the target Hamiltonian Used to represent the optimization objectives in smart apartment management, including energy distribution, equipment control, and user behavior optimization. The target Hamiltonian It is expressed as: in, is the Pauli-Z matrix, and They represent the parameters of device energy consumption, user behavior, and the relationship between devices during the optimization process;

[0036] 3) Time from Evolved to , the Hamiltonian gradually changes from the initial Hamiltonian Transition to the target Hamiltonian , the evolution form is: Evolution function It is generally a linear function, indicating that the system evolves gradually over time: exist When the system is in the initial Hamiltonian The ground state of exist When the system reaches the target Hamiltonian The ground state ;

[0037] 4) The quantum evolution process is numerically simulated through Trotter-Suzuki decomposition, which discretizes the entire evolution process and converts time Divide into small steps , the evolution operation is expressed as: in, is the time increment, is the discrete decomposition term of the Hamiltonian, Represents the time from the initial time 0 to time The evolution operation of Indicates the number of time steps.

[0038] The aforementioned data optimization method for a smart apartment, in step S4, includes:

[0039] 1) Generate a preliminary solution through the features extracted by the hypergraph neural network model , the preliminary solution is preliminarily trained by a classical optimization algorithm and used as the initial condition for quantum optimization;

[0040] 2) Quantum excited state evolution algorithm is used to preliminarily solve As the initial solution, a global search is performed, starting from the initial Hamiltonian The ground state gradually evolves to the target Hamiltonian The ground state of .

[0041] The aforementioned data optimization method for a smart apartment comprises, in step S5: 1) based on the optimization solution Dynamically adjust energy distribution within the apartment, reasonably distribute energy during peak hours, and reduce energy consumption;

[0042] 2) According to the optimization solution Control the operating status of devices in the apartment. Monitor the switch status and power consumption of devices in real time, and dynamically adjust the device operating mode according to current user needs and environmental conditions.

[0043] A computer device / equipment / system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0044] A computer-readable storage medium stores a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0045] Beneficial effects achieved by the present invention: The smart apartment data optimization method of the present invention utilizes a hypergraph neural network and a quantum adiabatic evolution algorithm (QAEA) to establish a mathematical model to achieve comprehensive optimization of energy management, environmental monitoring, equipment control, and user behavior analysis for smart apartments, thereby solving the problems of low computational efficiency, poor optimization effect, and poor real-time performance in multi-source heterogeneous data processing in smart apartment data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0047] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0048] Example 1

[0049] like Figure 1 As shown, this embodiment provides a data optimization method for a smart apartment, comprising the following steps:

[0050] S1. Collect multi-source heterogeneous data in smart apartments and build a hypergraph model;

[0051] S11. In the smart apartment, multi-source heterogeneous data is collected in real time through IoT sensors, smart devices and user terminals, including:

[0052] Environmental parameters: temperature, humidity, light, air quality, etc.;

[0053] Device status: switch status, operation mode, power consumption, etc. of smart home appliances;

[0054] User behavior: indoor positioning, activity trajectory, usage preferences, etc.

[0055] Energy consumption: real-time usage of electricity, water and gas.

[0056] S12. Preprocess the collected multi-source heterogeneous data, including data cleaning, missing value filling, normalization and feature extraction, to form a high-dimensional feature matrix;

[0057] S13. Constructing a hypergraph model of smart apartment based on preprocessed data ,in:

[0058] Node Set : Includes all devices, sensors and users;

[0059] Super edge set : Indicates a multivariate relationship, including multiple devices being used by the same user, or multiple sensors jointly monitoring the same environmental parameter;

[0060] S2. construct a hypergraph neural network model, which is used to extract high-order features from multi-source heterogeneous data of the smart apartment and analyze the multivariate relationships between various entities in the apartment, including rooms, devices, sensors, and users;

[0061] The hypergraph neural network model models the entities in the smart apartment and the multivariate relationships between entities as a hypergraph. The hypergraph neural network model can extract deeper relationships and features from the multivariate data in the smart apartment. In the data management of smart apartments, there is not only a one-to-one relationship between devices and sensors, but more complex multivariate relationships can be effectively represented by hypergraphs, thereby improving the accuracy and real-time performance of intelligent management decisions.

[0062] The input data of the hypergraph neural network model includes pre-processed multi-source heterogeneous data in the smart apartment, including, for example:

[0063] Environmental parameters: temperature, humidity, air quality, etc.;

[0064] Device status: device operation mode, power consumption, switch status, etc.

[0065] User behavior: indoor positioning, activity trajectory, usage preferences, etc.

[0066] Energy consumption: real-time electricity, water and gas usage.

[0067] The output data is a high-level feature representation of each entity in the smart apartment, which can be used for:

[0068] Anomaly detection: Detecting abnormal equipment, abnormal user behavior, or abnormal energy consumption;

[0069] User behavior analysis: Analyze user behavior patterns in the apartment to optimize device control and energy management strategies.

[0070] In the hypergraph neural network model, a multi-layer hypergraph convolutional network (HGCN) is designed to learn the high-order feature representation of nodes. The structure and formula of the multi-layer hypergraph convolutional network are as follows:

[0071] For Layer, Node The characteristic expression is:

[0072]

[0073] in: Is a node In the Hidden representation of layers; Contains nodes The hyperedge set of ;

[0074] It is a super edge The set of nodes contained; It is a super edge The weight can be set according to relevance or importance; and are learnable weight matrices and bias terms; is an activation function, such as ReLU or ELU, Indicates that node u is in The feature representation of the layer, node u is related to node Through the super edge Connected neighbor nodes, is the node currently being processed, and u is on the same hyperedge Other nodes in the computation node Feature updates.

[0075] According to the management task, the corresponding loss function is set, and the total loss function is:

[0076]

[0077] in: is the number of tasks; It is The loss function of each task; is the task weight; It is against loss; is the weight against loss;

[0078] The management tasks include:

[0079] Anomaly detection: Based on various sensor data, it detects equipment anomalies and energy usage anomalies in real time. By extracting high-level features of equipment operation data and environmental data, it identifies abnormal conditions and issues fault warnings in a timely manner to prevent losses and safety hazards caused by potential faults.

[0080] Equipment status feedback: Real-time monitoring of equipment operation status data, and adjustment of equipment control strategies based on equipment operation status data, so that the equipment can operate in the best state and reduce energy consumption.

[0081] In order to improve the generalization ability of the hypergraph neural network model, the generator and the discriminator Construct a generative adversarial network (GAN) with the adversarial loss as:

[0082] L adv (G,D)= E x∼ P data [logD(x)]+ E z∼ P z [log(1-D(G(z)))]

[0083] in, E x ∼ P data [ ⋅ ] Represents the distribution from real data The data sample extracted from , the data samples are the data actually observed in the smart apartment;

[0084] E z ∼ P z [ ⋅ ] Represents the latent variable Distribution The sample drawn from ;

[0085] Represents the discriminator for the sample The judgment result of

[0086] represents the generator from the latent variables Generated samples;

[0087] Through this adversarial loss function, the parameters of the generator and the discriminator can be optimized so that the generated samples are more similar to the real samples;

[0088] S3. Mapping the optimization problem in smart apartment management into a quantum Hamiltonian optimization problem; the optimization problem includes energy distribution optimization, including:

[0089] 1) Initial state setting: From the initial state To begin with, the initial state is determined by the initial Hamiltonian The ground state is determined, which represents the simplest configuration of the system;

[0090] 2) Construct a target Hamiltonian, the target Hamiltonian Used to represent the optimization objectives in smart apartment management, including energy distribution, equipment control, and user behavior optimization. The target Hamiltonian It is expressed as: in, is the Pauli-Z matrix, and They represent the parameters of device energy consumption, user behavior, and the relationship between devices during the optimization process;

[0091] 3) Time from Evolved to , the Hamiltonian gradually changes from the initial Hamiltonian Transition to the target Hamiltonian , the evolution form is: Evolution function It is generally a linear function, indicating that the system evolves gradually over time: exist When the system is in the initial Hamiltonian The ground state of When the system reaches the target Hamiltonian The ground state ;

[0092] 4) The quantum evolution process is numerically simulated through Trotter-Suzuki decomposition, which discretizes the entire evolution process and converts time Divide into small steps , the evolution operation is expressed as: in, is the time increment, is the discrete decomposition term of the Hamiltonian, Represents the time from the initial time 0 to time The evolution operation of Indicates the number of time steps.

[0093] S4. In order to further improve the computational efficiency, the classical optimization algorithm is used to generate a preliminary solution, and the quantum excited state evolution algorithm is used to perform a global search to obtain the optimal solution, and a collaborative optimization is performed. The process includes:

[0094] 1) Generate a preliminary solution through the features extracted by the hypergraph neural network model , the preliminary solution is preliminarily trained by a classical optimization algorithm and used as the initial condition for quantum optimization;

[0095] 2) Quantum excited state evolution algorithm is used to preliminarily solve As the initial solution, a global search is performed, starting from the initial Hamiltonian The ground state gradually evolves to the target Hamiltonian The ground state of .

[0096] The above collaborative optimization process improves the overall optimization efficiency of the system by shortening the quantum evolution time and utilizing the efficient convergence capability of the classical algorithm.

[0097] S5. The optimal solution obtained by collaborative optimization in step S4 Carry out energy dispatching and equipment control of smart apartments, including:

[0098] 1) Based on the optimization solution Dynamically adjust the energy distribution in the apartment, reasonably distribute energy during peak hours, and reduce energy consumption. During peak electricity consumption, the peak-shaving and valley-filling strategy is used to prioritize low-power devices and postpone the operation of high-energy consumption devices; during low-power consumption periods, high-energy consumption devices are started to maximize the use of low-cost energy. In addition, the scheduling plan is dynamically adjusted based on real-time feedback data to make energy use more efficient and flexible.

[0099] 2) According to the optimization solution Control the operating status of devices in the apartment. Monitor the on / off status and power consumption of devices in real time, and dynamically adjust the device operating mode according to current user needs and environmental conditions. For example, automatically turn off lighting and air conditioning when no one is in the room, or start related devices in advance according to the user's scheduled time to go home to improve the living experience. Communicate with devices through the Internet of Things protocol so that the device operating status can respond to optimization strategies in a timely manner. At the same time, monitor device abnormalities in real time and automatically adjust device status to prevent failures.

[0100] By real-time monitoring of equipment operating status and user feedback, and dynamically updating optimization models and strategies, the management system of smart apartments can maintain efficient and stable operation in a complex and changing environment.

[0101] In order to verify the effectiveness of the method of the present invention, the following simulation experiment was carried out.

[0102] Dataset: Simulates a smart apartment with 100 rooms, 200 devices, 50 sensors, and 100 users.

[0103] Time span: Data collection lasted for 30 days, with data recorded every hour.

[0104] Experimental comparison methods: traditional machine learning methods (random forest, SVM); ordinary deep learning methods (multi-layer perceptron, LSTM); graph neural network (GCN); the method of the present invention (HGN + QAEA).

[0105] Experimental results:

[0106] The comparison of anomaly detection accuracy is shown in Table 1.

[0107] Table 1. Comparison of common detection accuracy

[0108] method Detection accuracy (%) False alarm rate (%) Support Vector Machine (SVM) 85.6 10.5 LSTM 88.9 8.2 Graph Neural Networks (GCNs) 92.3 6.1 The present method (HGN + QAEA) 95.7 3.8

[0109] Before and after the implementation of the method of the present invention, a survey on user satisfaction was conducted, and the results are shown in Table 2.

[0110] Table 2 User satisfaction comparison table

[0111] Evaluation indicators Satisfaction before implementation (%) Satisfaction after implementation (%) Increase (%) Environmental comfort 78.5 92.3 +17.6 Service response speed 80.2 94.1 +17.3 Safety and reliability 85.0 96.5 +13.5 Overall satisfaction 81.2 94.3 +16.1

[0112] After implementing the method of the present invention, the energy consumption of the smart apartment is shown in Table 3.

[0113] Table 3 Comparison of energy consumption

[0114] Time period Energy consumption before implementation (kWh) Energy consumption after implementation (kWh) Energy saving (%) Peak 5000 4200 +16.0 Off-peak period 3500 3000 +14.3 Low period 2000 1800 +10.0 total 10500 9000 +14.3

[0115] It can be seen from the above experimental results that the method of the present invention has achieved remarkable results in many aspects of smart apartment data management: Enhanced anomaly detection capability: the accuracy of anomaly detection has been increased to more than 95%, and the false alarm rate has been reduced to less than 4%. Improved user satisfaction: the overall satisfaction has been increased by more than 16%, and users' evaluation of environmental comfort and service response speed has been greatly improved. Achieved energy conservation: the total energy consumption has been reduced by more than 14%, with obvious economic and environmental benefits.

[0116] This invention effectively solves the deficiencies in multi-source heterogeneous data processing, complex optimization problem solving and real-time performance in smart apartment management by introducing advanced technologies such as hypergraph neural network, quantum adiabatic evolution algorithm and classical-quantum collaborative optimization. It not only significantly improves the accuracy and efficiency of energy management, equipment control and user behavior analysis, but also provides technical support for smart building management and has broad application prospects.

[0117] Example 2

[0118] A computer device / equipment / system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0119] Example 3

[0120] A computer-readable storage medium stores a computer program / instruction, which implements the steps of the above method when executed by a processor.

[0121] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, systems and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0122] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable storage medium produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable data processing device to produce computer-implemented processing, so that the instructions executed on the computer or other programmable data processing device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0124] The above description is only a preferred embodiment of the present invention. It should be pointed out that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A data optimization method for a smart apartment, characterized in that: The following steps are involved: S1. Collect multi-source heterogeneous data in smart apartments and build a hypergraph model; S2. construct a hypergraph neural network model, which is used to extract high-order features from multi-source heterogeneous data of the smart apartment and analyze the multivariate relationships between various entities in the apartment, including rooms, devices, sensors, and users; S3. Mapping the optimization problem in smart apartment management into a quantum Hamiltonian optimization problem; the optimization problem includes energy distribution optimization; S4. Generate a preliminary solution using a classical optimization algorithm, use a quantum excited state evolution algorithm to perform a global search to obtain an optimal solution, and perform collaborative optimization; S5. The optimal solution obtained by collaborative optimization in step S4 Carry out energy scheduling and equipment control for smart apartments.

2. According to claim 1, a data optimization method for a smart apartment is characterized in that: In step S1, it includes: S11. In smart apartments, IoT sensors, smart devices and user terminals are used to collect multi-source heterogeneous data in real time, including environmental parameters, device status, user behavior and energy consumption; S12. Preprocessing the collected multi-source heterogeneous data; S13. Constructing a hypergraph model of smart apartment based on preprocessed data ,in: Node Set : Includes all devices, sensors and users; Super edge set : Indicates a multivariate relationship, including multiple devices being used by the same user, or multiple sensors jointly monitoring the same environmental parameter.

3. According to claim 1, a data optimization method for a smart apartment is characterized in that: In step S2, in the hypergraph neural network model, a multi-layer hypergraph convolutional network is designed to learn the high-order feature representation of nodes. The structure and formula of the multi-layer hypergraph convolutional network are as follows: For Layer, Node The characteristic expression is: ； in: Is a node In the Hidden representation of layers; Contains nodes The hyperedge set of ; It is a super edge The set of nodes contained; It is a super edge The weight of and are learnable weight matrices and bias terms; is the activation function, Indicates that node u is in The feature representation of the layer, node u is related to node Through the super edge Connected neighbor nodes, is the node currently being processed, and u is on the same hyperedge Other nodes in the computation node Feature updates.

4. The data optimization method for a smart apartment according to claim 3 is characterized in that: In step S2, the corresponding loss function is set according to the management task, and the total loss function is: ； in: is the number of tasks; It is The loss function of each task; is the task weight; It is against loss; is the weight against loss; The management tasks include anomaly detection and device status feedback.

5. According to the data optimization method of a smart apartment according to claim 4, it is characterized in that: In step S2, using the generator and the discriminator Construct a generative adversarial network with the adversarial loss as: ； in, Represents the distribution from real data The data sample extracted from , the data samples are the data actually observed in the smart apartment; Represents the latent variable Distribution The sample drawn from ; Represents the discriminator for the sample The judgment result of represents the generator from the latent variables Generated samples.

6. The data optimization method for a smart apartment according to claim 1, characterized in that: In step S3, it includes: 1) initial state setting: from the initial state To begin with, the initial state is determined by the initial Hamiltonian The ground state is determined by the system, which represents the simplest configuration of the system; 2) Construct a target Hamiltonian, the target Hamiltonian Used to represent the optimization objectives in smart apartment management, including energy distribution, equipment control, and user behavior optimization. The target Hamiltonian It is expressed as: ； in, is the Pauli-Z matrix, and They represent the parameters of device energy consumption, user behavior, and the relationship between devices during the optimization process; 3) Time from Evolved to , the Hamiltonian gradually changes from the initial Hamiltonian Transition to the target Hamiltonian , the evolution form is: ； Evolution function It is generally a linear function, indicating that the system evolves gradually over time: ； exist When the system is in the initial Hamiltonian The ground state of When the system reaches the target Hamiltonian The ground state ; 4) The quantum evolution process is numerically simulated through Trotter-Suzuki decomposition, which discretizes the entire evolution process and converts time Divide into small steps , the evolution operation is expressed as: ； in, is the time increment, is the discrete decomposition term of the Hamiltonian, Represents the time from the initial time 0 to The evolution operation of Indicates the number of time steps.

7. The data optimization method for a smart apartment according to claim 6, characterized in that: In step S4, it includes: 1) Generate a preliminary solution through the features extracted by the hypergraph neural network model , the preliminary solution is preliminarily trained by a classical optimization algorithm and used as the initial condition for quantum optimization; 2) Quantum excited state evolution algorithm is used to preliminarily solve As the initial solution, a global search is performed, starting from the initial Hamiltonian The ground state gradually evolves to the target Hamiltonian The ground state of .

8. According to the data optimization method of a smart apartment according to claim 7, in step S5, it comprises: 1) Based on the optimization solution Dynamically adjust energy distribution within the apartment, reasonably distribute energy during peak hours, and reduce energy consumption; 2) According to the optimization solution Control the operating status of devices in the apartment. Monitor the switch status and power consumption of devices in real time, and dynamically adjust the device operating mode according to current user needs and environmental conditions.

9. A computer device / equipment / system comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images