Solar acousto-optic equipment group cooperative control optimization method and system based on Internet of Things

Abstract

The invention provides a solar acousto-optic equipment group cooperative control optimization method and system based on the Internet of Things, and relates to the technical field of equipment control, and the method comprises the steps: analyzing the equipment data sequence characteristic correlation degree through a long-short-term memory network, and calculating an equipment trust value based on the equipment characteristic identification time sequence change; an equipment group is divided into a core node layer and a cooperative node layer, a verification sequence is established by using a Bloom filter, and a cross-domain resource allocation scheme is formed in combination with a long and short-term memory network. According to the invention, the cooperative control efficiency of the solar acousto-optic equipment group is improved, and the safety and adaptability of the system are enhanced.

Description

Technical Field

[0001] The present invention relates to device control technology, and in particular to a method and system for collaborative control optimization of a solar acousto-optic device group based on the Internet of Things. Background Art

[0002] With the rapid development of the Internet of Things (IoT) and the widespread adoption of renewable energy, solar-powered acoustic and optical devices have become a crucial component of smart cities and intelligent transportation. Solar-powered acoustic and optical device clusters typically consist of multiple distributed nodes, interconnected through IoT technology to collaboratively perform specific monitoring, warning, and information dissemination functions. Due to their clean, environmentally friendly, and self-sufficient energy supply, solar-powered acoustic and optical devices have found widespread application in scenarios such as road traffic, border security, and disaster warning. Currently, the control and management of solar-powered acoustic and optical device clusters primarily relies on centralized control systems, which centrally schedule and manage the devices through pre-set control strategies.

[0003] Existing technologies lack the ability to deeply mine and utilize historical equipment operation data, and are unable to effectively analyze the timing characteristics and correlation relationships of equipment operation status, resulting in a lack of pertinence and foresight in control strategies, and an inability to perform precise control and scheduling optimization based on the equipment's historical performance.

[0004] Traditional control methods usually use a unified control strategy to manage all devices, without considering the differences and hierarchies between devices. They fail to establish a hierarchical control mechanism based on device characteristics and trust, making it difficult to achieve differentiated collaboration and optimized resource configuration within a device group.

[0005] The existing solar acoustic and optical equipment group collaborative system lacks a secure and reliable mechanism for cross-domain data interaction and resource allocation. The security and integrity of data transmission are difficult to guarantee. At the same time, there is also a lack of intelligent allocation methods based on device characteristics and timing correlations, which limits the collaborative efficiency and adaptability of device groups in complex environments. Summary of the Invention

[0006] The embodiments of the present invention provide a method for collaborative control optimization of a solar acousto-optic device group based on the Internet of Things, which can solve the problems in the prior art.

[0007] A first aspect of an embodiment of the present invention provides a method for optimizing collaborative control of a solar-powered acousto-optical device group based on the Internet of Things, comprising: Mapping the position source data of the solar sound and light device group into multi-dimensional time series features to construct a device data sequence; using a long short-term memory network to analyze the feature correlation in the device data sequence, quantify the historical operation characteristics of the device, and mark the device feature identifier; Calculating the device trust value based on the temporal change trend of the device feature identifier, and dividing the solar acoustic and optical device group into a core node layer and a collaborative node layer according to the distribution characteristics of the device trust value; Decoupling the control instruction sequence of the core node layer based on the long short-term memory network, establishing a mapping relationship between the device feature identifier and the control instruction, and generating an initial control solution; establishing a verification sequence for the initial control solution based on the Bloom filter of the collaborative node layer, matching the device feature identifier to update the device trust value, and forming a final control solution; Based on the control sequence formed by the final control scheme, the intelligent control status of the solar acoustic and optical equipment group is analyzed in real time to construct cross-domain collaborative data; according to the Bloom filter, a hash mapping is established for the cross-domain collaborative data to generate security data, and the long short-term memory network is used to analyze the temporal characteristics of the security data and establish an association model with the device feature identifier to form a cross-domain resource allocation scheme, which is converted into device execution instructions via control sequence mapping.

[0008] The position source data of the solar sound and light equipment group is mapped into multi-dimensional time series features to construct an equipment data sequence. The feature correlation in the equipment data sequence is analyzed using a long short-term memory network to quantify the historical operation characteristics of the equipment. The equipment feature identifiers include: Constructing the state source data of the solar acousto-optic device group into a time series data sequence, constructing a quantum well-quantum dot composite structure, regulating the carrier transmission path based on the quantum confinement effect of the quantum well-quantum dot composite structure, mapping the time series data sequence into a quantum state, and adjusting the energy level distribution of the quantum well-quantum dot composite structure through quantum state engineering to form a quantum feature sequence; Utilizing the quantum tunneling effect to construct a quantum transmission channel, the quantum transmission channel includes a quantum tunneling barrier array, regulating carrier transport through the Coulomb blockade effect of the quantum tunneling barrier array, and generating a device data sequence based on the coupling of the quantum tunneling effect and the quantum confinement effect; A long short-term memory network is established to parse the device data sequence, the output characteristics of the long short-term memory network are correlated and mapped with the coherence characteristics of the quantum state, the historical operation characteristics of the device are quantified based on the coherence function of the quantum state, and a device feature identifier is generated.

[0009] Constructing a quantum transmission channel using the quantum tunneling effect, the quantum transmission channel includes a quantum tunneling barrier array, regulating carrier transport through the Coulomb blockade effect of the quantum tunneling barrier array, and generating a device data sequence based on the coupling of the quantum tunneling effect and the quantum confinement effect, including: Constructing a periodically arranged quantum tunneling barrier array, establishing a carrier wave function in the barrier region of the quantum tunneling barrier array using a gradient micro-nanostructure, and modulating an exponential wave vector parameter of the carrier wave function based on the gradient micro-nanostructure; Under the action of the carrier wave function, a Coulomb blockade effect control mechanism is constructed, surface plasmon resonance is used to enhance the coupling effect of the left and right potential barriers to generate a transmission coefficient, and the tunneling current corresponding to the carrier wave function is controlled based on the transmission coefficient; The Coulomb blockade effect control mechanism and the quantum confinement effect form a coupling system, and external field control is used to realize coupled state transmission and uncoupled state transmission of the coupling system, thereby generating a coupling enhancement factor. The tunneling current is modulated according to the coupling enhancement factor to generate a device data sequence.

[0010] Decoupling the control instruction sequence of the core node layer based on the long short-term memory network, establishing a mapping relationship between the device feature identifier and the control instruction, and generating an initial control plan include: A long short-term memory network is trained to establish a temporal association model of device features, and the control instruction sequence of the core node layer is decoupled based on the temporal association model. A gradient feedback adjustment mechanism is used to construct a mapping relationship between the device feature identifier and the control instruction. The gradient feedback adjustment mechanism adjusts the mapping weights in real time, and an initial control scheme is generated based on the iterative optimization results of the mapping weights.

[0011] Establishing a verification sequence for the initial control solution according to the Bloom filter of the collaboration node layer, matching the device feature identifier to update the device trust value, and forming a final control solution includes: Constructing a verification sequence for an initial control scheme based on a Bloom filter at the collaborative node layer, decomposing the initial control scheme into multiple control scheme components, mapping the control scheme components using a family of hash functions, inputting the mapping results into a neural symbolic reasoning system, and generating an optimized verification sequence based on the neural symbolic reasoning system combined with an expert knowledge base; constructing a causal inference-driven feature matching network based on the optimized verification sequence, performing causal correlation analysis between feature components and the optimized verification sequence based on the causal inference-driven feature matching network, and calculating the device trust value using a multi-agent collaborative evaluation mechanism; The device trust value is input into a transfer learning model, and the transfer learning model generates an optimized control strategy based on the basic control strategy and the new scenario characteristics. The optimized control strategy is trusted according to the device trust value to generate a final control solution.

[0012] Mapping the control scheme components using a family of hash functions, inputting the mapping results into a neural symbolic reasoning system, and generating an optimized verification sequence based on the neural symbolic reasoning system combined with an expert knowledge base includes: Decomposing a control scheme into components to obtain a plurality of control components, mapping the plurality of control components using a hash function family to obtain initial mapping features, constructing a component feature matrix based on the initial mapping features, determining a degree of correlation between components through similarity calculation, and adjusting the component feature matrix according to the degree of correlation to generate an enhanced mapping feature; Inputting the enhanced mapping features into a neural symbolic reasoning system, extracting information from the enhanced mapping features through the neural symbolic reasoning system to construct a feature priority sequence, reconstructing the enhanced mapping features based on the feature priority sequence to obtain reconstructed features, and analyzing the temporal variation patterns of the reconstructed features to generate temporal correlation features; Construct a knowledge fusion module, establish a mapping relationship between the time series correlation features and the expert knowledge base based on the knowledge fusion module, construct a feature expression space according to the mapping relationship, convert the time series correlation features into a standard feature representation in the feature expression space, and adjust the standard feature representation based on the balance factor to generate an optimized verification sequence.

[0013] Based on the control sequence formed by the final control scheme, the intelligent joint control status of the solar acoustic and optical device group is analyzed in real time to construct cross-domain collaborative data; and the cross-domain collaborative data is hashed and mapped to generate security data according to the Bloom filter, including: Performing chaotic mapping on the control sequence to obtain an iterative value of the device state, constructing a chaotic mapping model based on the iterative value of the device state, analyzing the dynamic change law of the device state parameter based on the chaotic mapping model, and generating a parameter mapping sequence based on the dynamic change law; Analyzing a state change trend of a device group according to the parameter mapping sequence, wherein the state change trend reflects a co-evolution process of the device group under the action of the parameter mapping sequence, determining the operational stability of the device group based on the co-evolution process, and dynamically adjusting a control strategy according to the operational stability using the chaotic mapping model; A hash verification space is constructed for the control strategy through the chaotic mapping model, a verification path of the control strategy is established in the hash verification space, and security verification information of the control strategy is transmitted along the verification path to generate security data.

[0014] A second aspect of an embodiment of the present invention provides a solar-powered acousto-optical device group collaborative control optimization system based on the Internet of Things, comprising: The first unit is used to map the position source data of the solar sound and light device group into multi-dimensional time series features to construct a device data sequence; use a long short-term memory network to analyze the feature correlation in the device data sequence, quantify the historical operation characteristics of the device, and mark the device feature identifier; The second unit is configured to calculate a device trust value based on a temporal variation trend of the device feature identifier, and divide the solar acoustic and optical device group into a core node layer and a collaborative node layer according to a distribution characteristic of the device trust value; The third unit is configured to decouple the control instruction sequence of the core node layer based on the long short-term memory network, establish a mapping relationship between the device feature identifier and the control instruction, and generate an initial control solution; establish a verification sequence for the initial control solution based on the Bloom filter of the collaborative node layer, match the device feature identifier to update the device trust value, and form a final control solution; The fourth unit is used to analyze the intelligent control status of the solar acoustic and optical equipment group in real time based on the control sequence formed by the final control scheme, and construct cross-domain collaborative data; establish a hash mapping for the cross-domain collaborative data according to the Bloom filter to generate security data, use the long short-term memory network to analyze the temporal characteristics of the security data and establish an association model with the device feature identifier to form a cross-domain resource allocation plan, and the cross-domain resource allocation plan is converted into device execution instructions through control sequence mapping.

[0015] According to a third aspect of an embodiment of the present invention, an electronic device is provided, including: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0016] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described above is implemented.

[0017] The beneficial effects of this application are as follows: By mapping the position source data of the solar acoustic and optical equipment group into multi-dimensional time series features and using the long short-term memory network to analyze the feature correlation in the equipment data sequence, the historical operation characteristics of the equipment can be accurately quantified and labeled, which improves the accuracy and timeliness of equipment feature identification and lays a data foundation for subsequent collaborative control.

[0018] Through the device trust value calculation and node level division mechanism, a hierarchical control architecture of the core node layer and the collaborative node layer was established, achieving precise decoupling and mapping of control instructions. At the same time, the Bloom filter was used to establish a verification sequence to optimize the control scheme, effectively improving the collaborative control efficiency and reliability of the solar acoustic and optical equipment group.

[0019] By constructing cross-domain collaborative data and using Bloom filters to establish hash maps to generate secure data, combined with long-short-term memory networks to analyze timing characteristics, secure and efficient communication and resource optimization and allocation between device groups are achieved, significantly enhancing the security, stability and adaptability of the system, and solving the problems of low collaborative efficiency and insufficient security of traditional control methods in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the process of the collaborative control optimization method of a solar acousto-optic device group based on the Internet of Things according to an embodiment of the present invention; Figure 2 A flowchart for constructing a quantum transmission channel according to an embodiment of the present invention; Figure 3 Generate an optimized verification sequence logic flow chart for the neural symbolic reasoning system of an embodiment of the present invention. DETAILED DESCRIPTION

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings 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 making creative efforts shall fall within the scope of protection of the present invention.

[0022] The technical solution of the present invention is described in detail below with reference to specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0023] Figure 1 FIG. 1 is a flow chart of a method for collaborative control optimization of a solar sound and light device group based on the Internet of Things according to an embodiment of the present invention. Figure 1 As shown, the method includes: Mapping the position source data of the solar sound and light device group into multi-dimensional time series features to construct a device data sequence; using a long short-term memory network to analyze the feature correlation in the device data sequence, quantify the historical operation characteristics of the device, and mark the device feature identifier; Calculating the device trust value based on the temporal change trend of the device feature identifier, and dividing the solar acoustic and optical device group into a core node layer and a collaborative node layer according to the distribution characteristics of the device trust value; Decoupling the control instruction sequence of the core node layer based on the long short-term memory network, establishing a mapping relationship between the device feature identifier and the control instruction, and generating an initial control solution; establishing a verification sequence for the initial control solution based on the Bloom filter of the collaborative node layer, matching the device feature identifier to update the device trust value, and forming a final control solution; Based on the control sequence formed by the final control scheme, the intelligent control status of the solar acoustic and optical equipment group is analyzed in real time to construct cross-domain collaborative data; according to the Bloom filter, a hash mapping is established for the cross-domain collaborative data to generate security data, and the long short-term memory network is used to analyze the temporal characteristics of the security data and establish an association model with the device feature identifier to form a cross-domain resource allocation scheme, which is converted into device execution instructions via control sequence mapping.

[0024] In an optional embodiment, the position source data of the solar sound and light device group is mapped into multi-dimensional time series features to construct a device data sequence; the feature correlation in the device data sequence is analyzed using a long short-term memory network to quantify the historical operation characteristics of the device, and the device feature identification is marked, including: Constructing the state source data of the solar acousto-optic device group into a time series data sequence, constructing a quantum well-quantum dot composite structure, regulating the carrier transmission path based on the quantum confinement effect of the quantum well-quantum dot composite structure, mapping the time series data sequence into a quantum state, and adjusting the energy level distribution of the quantum well-quantum dot composite structure through quantum state engineering to form a quantum feature sequence; Utilizing the quantum tunneling effect to construct a quantum transmission channel, the quantum transmission channel includes a quantum tunneling barrier array, regulating carrier transport through the Coulomb blockade effect of the quantum tunneling barrier array, and generating a device data sequence based on the coupling of the quantum tunneling effect and the quantum confinement effect; A long short-term memory network is established to parse the device data sequence, the output characteristics of the long short-term memory network are correlated and mapped with the coherence characteristics of the quantum state, the historical operation characteristics of the device are quantified based on the coherence function of the quantum state, and a device feature identifier is generated.

[0025] The acquisition of position source data from the solar-powered acousto-optic device cluster is accomplished through a distributed sensor network. The acquisition system consists of 48 nodes, each equipped with a power sensor, angle sensor, light sensor, and acoustic sensor, with a sampling frequency set to 100 Hz. The position source data includes device power data (range 0-500 watts, accuracy of 0.1 watt), azimuth data (range 0-360 degrees, accuracy of 0.5 degrees), light intensity data (range 0-2000 lumens, accuracy of 1 lumen), and acoustic output data (range 30-100 decibels, accuracy of 0.2 decibels). The data acquisition window is set to 24 hours, with each device generating 345,600 sampling points, for a total of 16,588,800 raw data points for the entire device cluster. The raw data is preprocessed by an edge computing unit, including outlier filtering (values ​​outside the normal range are marked as invalid), missing value interpolation (using linear interpolation), and data normalization (mapping to a range of -1 to 1). The processed position source data is organized into a time series data sequence. Each device forms an 8-dimensional time series vector (power, power change rate, angle, angle change rate, light intensity, light change rate, acoustic output, acoustic output change rate), and the sequence length is 86,400 points (downsampled at 10 Hz).

[0026] The quantum well-quantum dot composite structure was constructed by molecular beam epitaxy, which sequentially grew a 10 nm Al0.3Ga0.7As barrier layer, a 12 nm GaAs quantum well layer, a 2 nm Al0.3Ga0.7As spacer layer, and a 4 nm In0.25Ga0.75As quantum dot layer (dot density 8×10 10 centimeter -2 , average diameter 18 nanometers), and a 10-nanometer Al0.3Ga0.7As capping layer. The growth temperature is controlled at 580 degrees Celsius and the growth rate is 0.8 nanometers / minute. The quantum well layer provides a two-dimensional confined space with an energy level interval of 45 millielectronvolts; the quantum dot layer provides a zero-dimensional confined space with an energy level interval of 78 millielectronvolts between the ground state and the first excited state. The carrier transport path is regulated by an external electric field, and 30-nanometer Ti / Au electrodes are deposited on the top and bottom of the structure as top and back gates, respectively. The top gate voltage range is -1.0 to +1.0 volts, and the back gate voltage range is -5.0 to +5.0 volts. When the top gate voltage is +0.5 volts and the back gate voltage is +2.0 volts, the carrier wave function in the quantum well and the carrier wave function in the quantum dot overlap to the maximum extent, forming an optimal transmission channel, and the measured tunneling current is 125 nanoamperes; when the top gate voltage is -0.5 volts and the back gate voltage is -2.0 volts, the wave function overlap is minimal, and the tunneling current drops to 8 nanoamperes.

[0027] The time series data sequence vector state mapping adopts quantum coding technology. Each 8-dimensional time series vector is mapped to the quantum state of a 3-qubit system. The mapping relationship is: the power parameter is mapped to the θ angle (0-π) of the first qubit, the angle parameter is mapped to the φ angle (0-2π) of the first qubit, the light intensity is mapped to the θ angle of the second qubit, the acoustic output is mapped to the φ angle of the second qubit, and the four change rate parameters are respectively mapped to the θ angle and φ angle of the third qubit and the two global phases. In the actual mapping, the parameters of device D1 at time t = 1000 seconds were: power of 380 watts (normalized value 0.52) mapped to θ1 = 1.63 radians, direction angle of 175 degrees (normalized value -0.03) mapped to φ1 = 3.08 radians, light intensity of 650 lumens (normalized value 0.3) mapped to θ2 = 1.25 radians, acoustic output of 65 decibels (normalized value 0.2) mapped to φ2 = 3.77 radians, and four rate of change parameters (0.05, -0.02, 0.03, and 0.01) mapped to θ3 = 1.58 radians and φ3 = 3.02 radians, respectively, and two global phases of 0.03π and 0.01π. The resulting quantum state was verified by quantum state tomography, with a fidelity of 97.5%.

[0028] Quantum state engineering modulation utilizes microwave pulse technology. A microwave resonant cavity is integrated on the surface of a quantum well-quantum dot composite structure, with a resonant frequency of 8.5 GHz and a quality factor of 12,000. The microwave pulse train consists of π pulses (20 nanoseconds pulse width, 15 milliwatts power) and π / 2 pulses (10 nanoseconds pulse width, 7.5 milliwatts power). Precise manipulation of the quantum state is achieved by modulating the phase, amplitude, and timing of the microwave pulses. The microwave pulse train is generated by an arbitrary waveform generator and transmitted to the resonant cavity via a coaxial cable. Energy level distribution modulation is achieved by applying a perpendicular electric field with a strength range of 0-50 kilovolts per centimeter (kV / cm) in steps of 1 kV / cm. When the electric field strength is 35 kV / cm, the energy levels of the quantum well and quantum dot are optimally matched, and the cross-coupling strength between the energy levels reaches a maximum of 6.8 millielectronvolts. The quantum signature sequence is formed by continuously measuring the quantum state 864,000 times with a 100-millisecond interval, resulting in a complete quantum signature sequence. The coherence time of the sequence is 1.5 microseconds, which is much longer than the single measurement time (10 nanoseconds), ensuring the validity of the measurement.

[0029] The quantum tunneling barrier array is constructed using selective epitaxy and precision etching techniques to grow a 10-layer periodic structure on a GaAs substrate. Each layer contains a 3.5-nanometer Al0.4Ga0.6As barrier layer and a 5.0-nanometer GaAs well layer. The barrier layer has a band gap of 0.32 electron volts, and the well layer forms a quantum confinement structure with a confinement energy of 78 millielectron volts. The source and drain electrodes are prepared at the top and bottom of the array, respectively, using Ti / Au materials (thicknesses of 10 nanometers and 40 nanometers, respectively). Metal is deposited using electron beam evaporation technology with a vacuum degree of 5×10 -7 Torr, with a deposition rate of 0.5 angstroms per second. The gate is prepared on the side of the array using Cr / Au materials (thicknesses of 5 nanometers and 30 nanometers, respectively). The transmission characteristics of the barrier are regulated by precisely controlling the gate voltage. The Coulomb blockade effect is achieved by introducing quantum dots with a diameter of 20 nanometers into the fifth potential well. The quantum dots are made of InGaAs material and grown by molecular beam epitaxy at a growth temperature of 520 degrees Celsius for 60 seconds. The charging energy of the quantum dots is 18 millielectronvolts, which is much greater than the thermal energy (about 25 millielectronvolts at room temperature), ensuring the effectiveness of the Coulomb blockade effect.

[0030] Carrier transport control is achieved by adjusting the gate voltage, which ranges from -2.0 to +2.0 volts in 0.1 volt steps. At a gate voltage of -1.5 volts, the Coulomb blockade effect is maximized, reducing the tunneling current to a minimum of 5 nanoamperes. At a gate voltage of +1.5 volts, the Coulomb blockade effect is suppressed, and the tunneling current reaches a maximum of 350 nanoamperes. Periodic variations in the gate voltage generate a pulsed current train with a pulse width of 500 nanoseconds, a pulse interval of 1 microsecond, and a pulse height of 340 nanoamperes. The coupling of quantum tunneling and quantum confinement effects is achieved by adjusting the relative position of the quantum well-quantum dot composite structure and the quantum tunneling barrier array. The two structures are separated by 8 nanometers, forming a coupled system through Coulomb interaction. The coupling strength is determined by measuring the tunneling current enhancement factor. The maximum enhancement factor is 5.8, corresponding to an increase in tunneling current from 60 nanoamperes to 348 nanoamperes. The device data sequence is obtained by inputting a quantum signature sequence into the coupled system and measuring the output tunneling current sequence. This tunneling current sequence undergoes analog-to-digital conversion (12-bit resolution, 50 MHz sampling rate) to form a device data sequence with a length of 4,320,000 points.

[0031] The long short-term memory network was constructed using a deep learning framework. The network structure consists of an input layer (256 neurons), three LSTM hidden layers (128, 64, and 32 neurons, respectively), and an output layer (16 neurons). The input data is a device data sequence with a length of 1000 time steps, each containing a 256-dimensional feature vector. The network was trained with a batch size of 64, a learning rate of 0.001, and 200 training rounds. The training dataset consists of 30 days of historical operating data, totaling 27,000 samples; the validation set consists of 5 days of data, totaling 4,500 samples; and the test set consists of 3 days of data, totaling 2,700 samples. Early stopping was used during training to prevent overfitting, and training was terminated after 10 consecutive rounds of no improvement in the validation loss. The final model achieved an accuracy of 94.8%, a recall of 92.6%, and an F1 score of 93.7% on the test set.

[0032] The association mapping between LSTM output features and quantum state coherence properties utilizes an attention mechanism. A mapping relationship is established between the 32-dimensional output features of the last hidden layer of the LSTM and the elements of the density matrix of the quantum state. A self-attention layer is introduced during this mapping process, consisting of four attention heads, each with a dimension of 8. Attention weights are determined through training and reflect the correlation between different features and quantum state elements. Using data from device D1 as an example, the attention mechanism identified a high correlation between power fluctuation features and off-diagonal elements of the quantum state (correlation coefficient 0.87), and a high correlation between angular stability features and diagonal elements of the quantum state (correlation coefficient 0.92). This mapping relationship enables bidirectional conversion between classical data features and quantum state properties, with a conversion accuracy of 91.5%.

[0033] The quantum state coherence function is constructed based on a density matrix representation. The quantum state of each device is represented by an 8×8 density matrix. The diagonal elements of the matrix reflect the operating status of each functional module of the device, while the off-diagonal elements reflect the synergistic relationship between modules. The coherence function is calculated using the purity and von Neumann entropy of the density matrix. The purity ranges from 0.125 to 1, and the entropy ranges from 0 to 3. Normal device operation corresponds to a purity greater than 0.8 and an entropy less than 0.7; abnormal operation corresponds to a purity less than 0.6 and an entropy greater than 1.5. For device D5, for example, the average purity calculated from its 30-day operating data is 0.87 and the average entropy is 0.52, indicating good operating condition. By contrast, the average purity of device D12 is 0.58 and the average entropy is 1.74, indicating a fault.

[0034] The quantification of the historical operating characteristics of the equipment uses a sliding window technology, with the window size set to 24 hours (86,400 data points) and the sliding step size to 1 hour (3,600 data points). The four statistical characteristics of the quantum state coherence function, namely the mean, variance, peak value and trend, are calculated in each window. A total of 720 window feature vectors were generated during the 30-day operation period. The feature vectors were reduced in dimension through principal component analysis, and the first 8 principal components were retained, with the explained variance ratio reaching 94.3%. The feature vectors after dimensionality reduction were used for state recognition through a support vector machine classifier, and the equipment status was divided into three categories: "normal", "sub-healthy" and "faulty". The classification accuracy reached 96.7%, the false alarm rate was less than 2.5%, and the missed alarm rate was less than 1.8%.

[0035] The device signature is generated using hash coding technology. The quantized device historical operation feature vector is input into the SHA-256 hash algorithm to generate a 256-bit hash value as the basis for the device signature. The hash value is then encoded and converted into a 64-character string, such as "7A3B5C2D9E8F4G6H". The first 16 characters of the signature represent basic device information (model, location, etc.), the middle 32 characters represent the device's historical operation characteristics, and the last 16 characters represent the current status information. The signature update frequency is set to once every 24 hours, and historical versions are retained during updates to form a signature chain. By comparing the signatures of adjacent time points, the degree of change in the device status can be calculated. A change greater than 30% triggers an alarm mechanism. Actual applications have shown that this signature identification technology can predict equipment failures 72 hours in advance with a prediction accuracy of 89.5%, effectively reducing unexpected downtime and maintenance costs.

[0036] In an optional embodiment, a quantum transmission channel is constructed using the quantum tunneling effect, the quantum transmission channel includes a quantum tunneling barrier array, carrier transport is regulated by the Coulomb blockade effect of the quantum tunneling barrier array, and a device data sequence is generated based on the coupling of the quantum tunneling effect and the quantum confinement effect, including: Constructing a periodically arranged quantum tunneling barrier array, establishing a carrier wave function in the barrier region of the quantum tunneling barrier array using a gradient micro-nanostructure, and modulating an exponential wave vector parameter of the carrier wave function based on the gradient micro-nanostructure; Under the action of the carrier wave function, a Coulomb blockade effect control mechanism is constructed, surface plasmon resonance is used to enhance the coupling effect of the left and right potential barriers to generate a transmission coefficient, and the tunneling current corresponding to the carrier wave function is controlled based on the transmission coefficient; The Coulomb blockade effect control mechanism and the quantum confinement effect form a coupling system, and external field control is used to realize coupled state transmission and uncoupled state transmission of the coupling system, thereby generating a coupling enhancement factor. The tunneling current is modulated according to the coupling enhancement factor to generate a device data sequence.

[0037] like Figure 2 As shown, the method includes: Prepare a silicon (100) crystal substrate, use acetone, isopropanol and deionized water to ultrasonically clean for 10 minutes, blow dry with nitrogen, and then anneal in a 580°C annealing furnace for 30 minutes to remove the surface oxide layer. Use molecular beam epitaxy equipment in an ultra-high vacuum environment (1×10 -10 A 2-nanometer GaAs buffer layer was first grown, followed by alternating AlGaAs barrier and GaAs well layers. The AlGaAs barrier layer growth parameters were: temperature 580°C, V / III ratio 20:1, growth rate 1.0 nm / min, and growth time 210 seconds, resulting in a 3.5-nanometer thickness. The GaAs well layer growth parameters were: temperature 600°C, V / III ratio 15:1, growth rate 1.2 nm / min, and growth time 250 seconds, resulting in a 5.0-nanometer thickness. The barrier and well layer growth cycles were repeated 10 times to form a periodically arranged quantum tunneling barrier array. After growth, the layers were annealed at 185°C for 15 minutes to smooth the interface. X-ray diffractometer measurements confirmed that the barrier array had a period error of less than 0.2 nm.

[0038] 950K PMMA electron beam photoresist was spin-coated on the surface of the quantum tunneling barrier array at 4000 rpm for 45 seconds, forming a 180-nm thick photoresist layer. After a soft bake at 180°C for 2 minutes, exposure was performed using an electron beam lithography system with a dose of 280 μC / cm², an energy of 30 keV, and a beam diameter of 8 nm. Development was performed at 21°C for 45 seconds using a developer solution of isopropyl alcohol:methyl isobutyl ketone (MIBK) in a ratio of 1:3, resulting in a pattern with a line width of 50 nm. A reactive ion etching system was used with a gas ratio of BCl₃:Ar in a ratio of 1:5 and a power of 75 W. The etching time was divided into three steps: the first step was 15 seconds to etch a 5-nm depth, the second step was 30 seconds to etch to a 15-nm depth, and the third step was 20 seconds to etch to a 25-nm depth. A gradient depth profile was achieved by adjusting the etching time. The remaining photoresist was removed using an acetone ultrasonic cleaner for 5 minutes, followed by a 30-second isopropyl alcohol rinse and nitrogen blow-drying. The gradient micro-nanostructure was measured using atomic force microscopy, confirming that the depth increased linearly from 5 nanometers at the edge to 25 nanometers in the center, the lateral size was 500 nanometers, and the surface roughness was controlled below 0.5 nanometers.

[0039] Titanium / gold electrodes with a thickness of 50 nanometers were deposited at both ends of the gradient micro-nanostructure using electron beam evaporation with deposition rates of 0.2 Å / s and 1.0 Å / s, respectively. Photolithography and wet etching were used to define the electrode shape to form the source and drain electrodes. 100 nanometers of aluminum were deposited on the back of the structure as a back-gate electrode. A microneedle stage was used to connect the electrodes, and a bias voltage was applied using a semiconductor parameter analyzer, starting from 0.5 volts and increasing in steps of 0.1 volts to 2.0 volts. The tunneling current was measured at each voltage point, and the current-voltage curve was plotted. Based on the measurement data, it was determined that at a bias of 1.2 volts, the carrier wave function performed best in the barrier region, with an exponential wave vector parameter of 2.3×10 9 rice -1 Through the temperature control system, measurements were performed every 20 Kelvin in the range of 10 Kelvin to 300 Kelvin, the temperature dependence of the exponential wave vector parameters was recorded, and a temperature compensation curve was established.

[0040] Electron beam lithography and wet etching were used to form a 20-nanometer-diameter pit in the center of the quantum tunneling barrier array. Indium gallium arsenide quantum dots were grown in the pit using molecular beam epitaxy (MBE) with growth parameters of 520°C, a V / III ratio of 25:1, and 85 seconds, resulting in a quantum dot height of 8 nanometers. A 2.5-nanometer-thick aluminum arsenide capping layer was grown above the quantum dots using parameters of 550°C, a V / III ratio of 18:1, and 150 seconds. A 30-nanometer-thick silicon dioxide insulating layer was deposited on top of the structure using plasma-enhanced chemical vapor deposition (PECVD) with a 1:20 ratio of SiH₄ and N₂O at 60 watts at 300°C for 180 seconds. A 40-nanometer titanium / gold top gate electrode was deposited on top of the silicon dioxide. Coulomb blockade characteristics were measured using a semiconductor parameter analyzer. The top gate voltage was swept from -1.0 V to +1.0 V in 10 mV steps, with the source-drain voltage fixed at 10 mV. The Coulomb blockade step in the current-gate-voltage curve was observed, confirming a step height of 15 nanoamperes and a step width of 75 mV.

[0041] HSQ electron beam photoresist was spin-coated on the surface of the quantum tunneling barrier array at a speed of 6000 rpm for 60 seconds to form a photoresist layer with a thickness of 100 nm. After soft baking at 90 degrees Celsius for 2 minutes, it was exposed using an electron beam lithography system with a dose of 850 microcoulombs / square centimeter and an energy of 50 keV. TMAH developer was used to develop at 40 degrees Celsius for 60 seconds to obtain a nano-grating structure with a period of 150 nm and a line width of 60 nm. A 30 nm thick silver film was deposited using an electron beam evaporation coating device with a deposition rate of 0.5 angstroms / second and a vacuum degree of 5×10 -7Acetone ultrasonic peeling was performed for 2 minutes to form a silver nanostripe array. A tunable laser was used to provide a light source with a wavelength of 650 nanometers and a power of 30 milliwatts per square centimeter, which was guided to the sample surface through a fiber coupler. The surface plasmon field enhancement effect was measured using a near-field scanning optical microscope, confirming that the field enhancement factor was 15 times. A phase-locked amplifier was used to measure the tunneling current under different illumination conditions, and the change in the transmission coefficient from 0.007 to 0.085 was recorded.

[0042] A special quantum well structure was designed in the quantum tunneling barrier array, increasing the thickness of the 5th and 8th wells to 8 nanometers. The growth parameters were adjusted to: temperature 600 degrees Celsius, V / III ratio 15:1, growth rate 1.2 nanometers / minute, and growth time 400 seconds. When preparing the dual-gate structure, the back gate was doped with a concentration of 5×10 18 centimeter -3 The device was constructed using an n-type silicon substrate, a 30-nanometer-thick platinum film for the top gate, and a 70-nanometer distance between the two gates. Connected to a semiconductor parameter analyzer, the top gate voltage range was set to -1.0 to +1.0 volts, and the back gate voltage range was set to -5.0 to +5.0 volts. The two gate voltages were varied in steps of 50 millivolts and 100 millivolts, respectively, forming a gate voltage matrix. Tunneling currents were measured for each gate voltage combination, and a three-dimensional current response graph was plotted. The coupled operating point was determined to be +0.8 volts for the top gate and +3.5 volts for the back gate, while the uncoupled operating point was -0.5 volts for the top gate and -2.0 volts for the back gate. A pulse generator was used to generate gate voltage pulses with a width of 20 nanoseconds, and the measurement system response time was 47 nanoseconds.

[0043] A vector network analyzer (VNA) was used to measure the system's RF response at various gate voltage combinations over a frequency range of 100 MHz to 10 GHz. The system's equivalent circuit model was extracted using S-parameters, and the coupling enhancement factor (CEF) at various operating points was calculated. With the top gate voltage fixed at +0.5 volts, the back gate voltage was varied from -4.0 volts to +4.0 volts in 0.5-volt steps, and the corresponding tunneling current and CEF were measured. A curve plotting the CEF versus tunneling current confirmed that the tunneling current was 30 nanoamperes at a CEF of 1.0, 78 nanoamperes at a CEF of 2.0, 230 nanoamperes at a CEF of 4.0, and 385 nanoamperes at a CEF of 5.2. The least-squares method was used to fit the curves and establish a quantitative relationship model between the CEF and tunneling current, determining the model parameters. Using a magnetic field measurement system, a perpendicular magnetic field was applied from 0 to 2 Tesla in 0.1-Tesla steps at a constant temperature of 4.2 Kelvin. The CEF was measured under various magnetic fields to form a magnetic field manipulation dataset.

[0044] A high-speed data acquisition card was used with a sampling rate of 50 MHz, a resolution of 12 bits, and an input range of ±500 nanoamperes. Specific gate voltage waveforms were designed, including: square wave (frequency 1 MHz, amplitude ±0.5 volts), sine wave (frequency 2 MHz, amplitude ±0.3 volts), and random sequence (bandwidth 5 MHz). An arbitrary waveform generator was used to generate the gate voltage waveform, which was connected to the top gate and back gate of the device via a coaxial cable. A current preamplifier was used to amplify the tunneling current signal with a gain of 10. 7 Volt / ampere, bandwidth is 10 MHz. Use an oscilloscope to monitor the amplified current signal in real time and transfer the data to a computer. Write a data processing program to perform a 50-point moving average filter on the collected raw data to remove high-frequency noise, and then perform maximum-minimum normalization processing to map the data to the range of 0-1. Binarize the normalized data according to the preset threshold (0.5) and convert it into a binary sequence. Use 8B / 10B encoding to encode the binary sequence to improve transmission reliability. Divide the encoded data into 1024-bit data packets, and add a 32-bit cyclic redundancy check code to each data packet. Use a serial data interface to transmit the data packet to a storage device to form a device data sequence in a standard format. Use a bit error rate tester to measure the quality of the data sequence. The bit error rate is less than 10 at room temperature. -9 , the signal-to-noise ratio is greater than 45 decibels. The test was repeated under different temperature conditions (15 to 45 degrees Celsius, step size 5 degrees Celsius) to establish a relationship model between temperature and data quality.

[0045] A test platform was established, including a temperature control system (accuracy ±0.1 degrees Celsius), a power supply system (accuracy ±1 millivolt), a light source system (wavelength adjustable ±1 nanometer), and a measurement system (accuracy ±0.1 nanoamperes). The device power consumption was measured at four temperatures: 15, 25, 35, and 45 degrees Celsius, with values ​​of 2.15, 1.95, 2.08, and 2.65 milliwatts, respectively. A pseudo-random bit sequence generator was used to generate test data with a data length of 10 7 The measured transmission rates were 21, 20.5, 19.8 and 18.2 megabits per second. A noise generator was used to generate external noise with intensities of 5, 10, 15 and 20 decibels, and the accuracy of the measured data was 99.2%, 98.1%, 96.7% and 95.3% respectively. The quantum state storage time was measured and recorded as 2.8, 2.5, 2.1 and 1.7 microseconds. By comparing the test data with theoretical expectations, the system performance was evaluated, the performance indicators were calculated, and a complete technical evaluation report was formed. The test results show that the quantum transmission channel constructed by this method is superior to traditional semiconductor devices in terms of power consumption, data transmission rate, anti-interference ability and quantum state storage time, verifying the technical advantage of the coupling of quantum tunneling effect and quantum confinement effect.

[0046] In an optional embodiment, decoupling the control instruction sequence of the core node layer based on the long short-term memory network, establishing a mapping relationship between the device feature identifier and the control instruction, and generating an initial control solution includes: A long short-term memory network is trained to establish a temporal association model of device features, and the control instruction sequence of the core node layer is decoupled based on the temporal association model. A gradient feedback adjustment mechanism is used to construct a mapping relationship between the device feature identifier and the control instruction. The gradient feedback adjustment mechanism adjusts the mapping weights in real time, and an initial control scheme is generated based on the iterative optimization results of the mapping weights.

[0047] Collect characteristic data during device operation, including operating status parameters, environmental parameters, and user operation behavior data. This characteristic data is timestamped to form a time series dataset. For example, for a smart air conditioning system, the collected characteristic data includes indoor temperature changes (e.g., from 26°C to 22°C), humidity levels (e.g., from 65% to 50%), user-set temperatures (e.g., set to 22°C), and corresponding control commands (e.g., starting cooling mode and adjusting the fan speed to medium).

[0048] The collected time series datasets undergo preprocessing, including data cleaning, standardization, and feature extraction. Data cleaning removes outliers and missing values, such as filtering out obvious errors like the occasional -50°C temperature sensor reading. Standardization unifies feature data of different dimensions to the same scale, such as mapping temperature and humidity data to a range of 0-1. Feature extraction extracts more representative features from the raw data, such as calculating derived features such as the temperature change rate and user adjustment frequency.

[0049] The preprocessed data is used to train a long short-term memory (LSTM) network model, which employs a four-layer architecture: an input layer receives device feature vectors; an LSTM layer captures temporal dependencies; a fully connected layer performs feature mapping; and an output layer generates control command predictions. During training, a batch size of 64 packets is used to optimize network parameters via backpropagation. For example, the training data includes operating data for an air conditioning system at different times of the week. The model learns the user's preferred control strategy under specific temperature variation patterns.

[0050] The trained long-short-term memory network is used to decouple the control instruction sequences at the core node layer. The decoupling process first identifies key transition points in the control sequence—points where significant changes in the control strategy occur. For example, when the indoor temperature drops from 25°C to 23°C, the air conditioner switches from high to medium speed. The system analyzes the state changes before and after these transition points, extracting the triggering conditions for the state transitions and the corresponding control instruction changes.

[0051] The mapping between the decoupled control instruction sequence and the device feature identifiers is established through a gradient feedback regulation mechanism. This mechanism sets the initial mapping weight matrix W, with dimensions m×n, where m represents the device feature dimension and n represents the control instruction dimension. For example, for an intelligent lighting system with eight feature parameters and five control instructions, an 8×5 weight matrix is ​​initialized, with an initial value of 0.1.

[0052] The gradient feedback adjustment process iteratively adjusts the mapping weights in real time. In each iteration, the system inputs the current device feature vector X and multiplies it by the weight matrix W to obtain the predicted control command Y'. Y' is compared with the actual optimal control command Y, and the error E is calculated. Based on the error E, the gradient direction is calculated, and the weight matrix W is updated according to the learning rate α (typically set to 0.01). For example, when the error between the predicted control command and the actual control command is large, the weights between the corresponding features and the command are significantly adjusted.

[0053] The iterative optimization process continues until the termination condition is met: the error E falls below a preset threshold (e.g., 0.05) or the maximum number of iterations (e.g., 1000) is reached. In a real-world application, for example, the optimization process for a smart home system achieved convergence after 786 iterations, with the error dropping to 0.048.

[0054] Based on the optimized mapping weights, an initial control plan is generated. This plan includes a detailed mapping table of device characteristics and control commands, as well as control strategies for different scenarios. For example, when the indoor brightness is detected to be below 100 lux and someone is present, the intelligent lighting system automatically turns on the main light and adjusts the brightness to 350 lux. When the indoor temperature is above 28°C and the humidity is greater than 70%, the air conditioning system activates dehumidification mode and sets the target temperature to 25°C.

[0055] To verify the effectiveness of the control scheme, the system was tested in a simulated environment. The test data included 50 sets of device states and corresponding optimal control instructions for different scenarios. The generated initial control scheme was used to control these scenarios, and the degree of control accuracy compared to the optimal scheme was measured. The test results showed that the control scheme generated by this method achieved an average accuracy of 92.7% across various scenarios, effectively adapting to changes in device characteristics and making corresponding control adjustments.

[0056] Through the above method, the system successfully realized the decoupling of the core node layer control instruction sequence based on the long short-term memory network, established the mapping relationship between device feature identification and control instructions, and generated high-quality initial control schemes, providing an effective solution for the automated control of smart devices.

[0057] In an optional embodiment, establishing a verification sequence for the initial control solution based on the Bloom filter of the collaboration node layer, matching the device feature identifier to update the device trust value, and forming a final control solution includes: Constructing a verification sequence for an initial control scheme based on a Bloom filter at the collaborative node layer, decomposing the initial control scheme into multiple control scheme components, mapping the control scheme components using a family of hash functions, inputting the mapping results into a neural symbolic reasoning system, and generating an optimized verification sequence based on the neural symbolic reasoning system combined with an expert knowledge base; constructing a causal inference-driven feature matching network based on the optimized verification sequence, performing causal correlation analysis between feature components and the optimized verification sequence based on the causal inference-driven feature matching network, and calculating the device trust value using a multi-agent collaborative evaluation mechanism; The device trust value is input into a transfer learning model, and the transfer learning model generates an optimized control strategy based on the basic control strategy and the new scenario characteristics. The optimized control strategy is trusted according to the device trust value to generate a final control solution.

[0058] The Bloom filter construction at the collaborative node layer uses a bitmap storage structure with a bitmap size of 8192 bits and 20 independent hash functions. The Bloom filter construction process extracts a 64-bit device identification code from each device in the collaborative node layer. For example, the identification code for device D1 is "7A3B5C2D9E8F4G6H." This code is sequentially input into 20 hash functions, resulting in 20 hash values ​​ranging from 0 to 8191. The corresponding positions in the Bloom filter bitmap are marked as 1. For example, if the hash values ​​are 134, 2045, 3721, etc., then bits 134, 2045, and 3721 in the bitmap are set to 1. This process is repeated for all 48 devices in the collaborative node layer to complete the Bloom filter initialization. The initial control scheme includes eight parameters: voltage control, power distribution, angle adjustment, and lighting intensity. Each device has specific parameter values ​​for these eight dimensions.

[0059] Taking power allocation as an example, the parameters for device D1 are 385 watts, those for device D2 are 412 watts, those for device D3 are 367 watts, and so on, for a total of 48 power parameter values. When constructing the verification sequence, each parameter of the initial control solution is linked to the corresponding device identifier and input into a Bloom filter for verification. If the verification pass rate is less than 90%, the control solution is considered to require adjustment; if the pass rate is between 90% and 98%, it is marked as a solution for optimization; if the pass rate exceeds 98%, it is marked as a highly confident solution. In actual testing, the verification pass rate of the initial control solution was 94.6%, and it was marked as a solution for optimization.

[0060] Functional domain division splits the control scheme into eight functional components, including voltage control, power distribution, and angle adjustment. Parameter clustering is performed within each functional component, dividing the 48 devices into 5-8 cluster groups based on parameter similarity. Taking the power distribution component as an example, the power parameters of the 48 devices are clustered into 6 groups: the 350-380 watt group contains 12 devices, the 381-410 watt group contains 15 devices, the 411-440 watt group contains 9 devices, the 441-470 watt group contains 7 devices, the 471-500 watt group contains 3 devices, and the 300-349 watt group contains 2 devices. For each cluster group of each functional component, statistical characteristics such as parameter mean, standard deviation, and change trend are calculated.

[0061] The voltage control component has a mean parameter of 220 volts and a standard deviation of 8.6 volts; the power allocation component has a mean parameter of 402 watts and a standard deviation of 42.5 watts; and the angle adjustment component has a mean parameter of 175 degrees and a standard deviation of 35.2 degrees. The hash function family includes eight algorithms, including SHA-256, MD5, FNV-1a, and Murmur3, each responsible for processing a specific functional component. The power allocation component is processed using SHA-256 to generate a 256-bit hash value, "A7F3D9E2B5C8G1H6J4K2L8M3N9P5Q7." The angle adjustment component is processed using MD5 to generate a 128-bit hash value, "B8D4F2H6J1L3N5P7." After processing all eight functional components, a hash mapping result with a total length of 1536 bits is obtained.

[0062] The neural symbolic reasoning system uses a hybrid architecture combining a feedforward deep network with a symbolic reasoning engine. The neural network consists of a five-layer fully connected network. The input layer has 1,536 neurons corresponding to the hash map results, three hidden layers with 1,024, 512, and 256 neurons, respectively, and an output layer with 128 neurons. The training dataset consists of 10,000 sets of historical control cases, each containing input features and corresponding expert evaluation results. Training uses a batch size of 64, a learning rate of 0.0008, and 500 training rounds, achieving a validation accuracy of 93.8%. The symbolic reasoning engine includes 620 inference rules, such as "If the power exceeds 450 watts and the run time exceeds 6 hours, reduce the power to 400 watts" and "If the ambient temperature is above 38 degrees Celsius and the power exceeds 420 watts, increase cooling and ventilation."

[0063] The expert knowledge base contains 1,500 historical cases and 800 empirical rules, covering normal operating parameter ranges, fault diagnosis methods, and optimized control strategies. Examples include "When continuous rainy weather exceeds 72 hours, the energy storage system charging power should be reduced by 30% to prioritize power supply to core equipment"; and rules such as "In the sound and light collaborative working mode, for every 10% decrease in lighting intensity, sound clarity must be increased by 5% to maintain the overall experience." The neural symbolic reasoning system combines hash mapping results, inference rules, and knowledge base cases for analysis to generate a 2048-bit optimized verification sequence, consisting of original and supplementary verification information. The optimized verification sequence sets verification thresholds for each functional component, such as ±8% for the power allocation component and ±12 degrees for the angle adjustment component.

[0064] The causal inference-driven feature matching network is constructed using a combination of graph neural networks and causal inference models. The network architecture consists of six graph convolutional layers, each with 64 convolution kernels of 3×3 size and a stride of 1. The network inputs are the optimized verification sequence and device feature identifiers. When constructing the feature graph, each node represents a device or functional component, and the edge weights between nodes represent the strength of the association. Causal relationship analysis utilizes counterfactual reasoning to analyze each control parameter intervention and assess its impact on overall system performance.

[0065] For example, the power parameter of device D1 was virtually adjusted from 385 watts to 425 watts to analyze the system response; it was then adjusted to 345 watts to compare the system behavior under different interventions. Through multiple intervention experiments, the degree of causal influence of the parameters was determined. Causal analysis of device D1's signature and the power allocation verification sequence yielded a causal correlation strength of 0.89; the causal correlation strength with the angle adjustment verification sequence was 0.67; and the causal correlation strength with the lighting intensity verification sequence was 0.82. After completing the causal analysis of all 48 devices and eight functional components, a 48×8 causal correlation matrix was formed. Control weights were set based on the causal correlation strength: a control weight of 0.4 for correlation strengths greater than 0.85, 0.3 for correlation strengths between 0.7 and 0.85, 0.2 for correlation strengths between 0.5 and 0.7, and 0.1 for correlation strengths less than 0.5.

[0066] The multi-agent collaborative evaluation mechanism deploys 48 agent nodes, each corresponding to a physical device. The internal architecture of each agent comprises a data acquisition module, a state assessment module, a decision generation module, and a communication coordination module. The data acquisition module collects device operating data every five seconds, including 12 parameters such as voltage, current, power, temperature, light intensity, and sound loudness. The state assessment module calculates the device health based on this data, with a health score ranging from 0 to 100. The decision generation module generates control recommendations based on the health and causal association matrix. The communication coordination module is responsible for exchanging information with neighboring agents. Agent communication utilizes an encrypted P2P protocol, with a communication radius of 30 meters and a frequency of every 15 seconds.

[0067] Each agent forms a communication network with 5-8 surrounding agents, exchanging control recommendations and status information. The collaborative evaluation process proceeds through 12 rounds, with device trust values ​​updated after each round. Trust value calculations consider the device's historical performance (weighted 0.3), current status (weighted 0.5), and neighbor evaluations (weighted 0.2). Device historical performance is calculated based on the past 30 days of operational data, including failure rates, response speeds, and energy efficiency metrics. Current status is calculated based on 12 parameters collected in real time. Neighbor evaluations are the average ratings of surrounding agents for the device. Initially, the trust values ​​of all 48 devices were set to 0.8. After 12 rounds of collaborative evaluation, the trust values ​​ranged from 0.62 to 0.96. Eleven devices were classified as high-trust devices with a trust value above 0.9; 28 devices were classified as medium-trust devices with a trust value between 0.75 and 0.9; and nine devices were classified as low-trust devices with a trust value below 0.75.

[0068] The transfer learning model uses a multi-domain adaptation architecture, consisting of a feature extraction network, a domain adaptation layer, and a task output network. The feature extraction network uses a ResNet structure with 25 residual blocks; the domain adaptation layer uses adversarial training and includes a domain discriminator and a feature generator; and the task output network uses a three-layer fully connected network. Model pre-training uses 800 sets of historical control data, including optimal control strategies under various environmental conditions. The basic control strategy dataset contains control parameters under source domain environmental conditions (temperature 15-30°C, humidity 40-70%, and light intensity 500-800 lux). New scene features contain sensor data under target domain environmental conditions (temperature 25-40°C, humidity 60-90%, and light intensity 300-600 lux). The domain adaptation process achieves knowledge transfer by minimizing the difference in feature distribution between the source and target domains.

[0069] Taking temperature adaptation as an example, when the ambient temperature rose from 25°C to 38°C, the transfer learning model automatically adjusted the power parameter from an average of 402 watts to an average of 358 watts, a 10.9% decrease; the lighting intensity from an average of 750 lumens to 620 lumens, a 17.3% decrease; and the sound gain from an average of 0.75 to 0.85, an improvement of 13.3%. Evaluation of the model's adaptation effectiveness showed that the control strategy achieved an accuracy of 87.6% under unseen environmental conditions, a significant improvement over the 68.2% accuracy of directly applying the source domain strategy.

[0070] The trust mapping process combines the device trust value with the optimized control strategy to generate the final control solution. This mapping uses a piecewise function, setting different processing strategies for different trust value ranges. High-trust devices (trust value > 0.9) use a fully optimized control strategy with a parameter adjustment range of ±5%. Medium-trust devices (trust value 0.75-0.9) use a partially constrained control strategy with a parameter adjustment range of ±15% and increased verification frequency. Low-trust devices (trust value < 0.75) use a strictly constrained control strategy with a parameter adjustment range of ±30%, and increased real-time monitoring and redundant backup.

[0071] In actual applications, the trust value of device D5 was 0.94 (high confidence), and its power parameter was optimized from 425 watts to 410 watts, a 3.5% adjustment. The trust value of device D12 was 0.83 (medium confidence), and its power parameter was adjusted from 390 watts to 340 watts, a 12.8% adjustment. The trust value of device D37 was 0.68 (low confidence), and its power parameter was limited from 450 watts to 350 watts, a 22.2% adjustment. The monitoring frequency was also increased from the standard once per minute to once every 10 seconds. The final control solution includes 8-dimensional control parameters for 48 devices, totaling 384 parameters. All parameters were adjusted through trust mapping to match the actual capabilities of the devices, ensuring the overall stable operation of the system.

[0072] Through a 45-day test verification in an experimental environment consisting of 48 solar-powered sound and light devices, the performance differences between the traditional control method and this method were compared. The test results show that after adopting this method, the energy consumption of the equipment group was reduced by 21.6%, from an average of 15.8 kWh per day to 12.4 kWh; the system failure rate was reduced by 76.3%, from an average of 12 times per week to 2.8 times; the average response time was shortened from 210 milliseconds to 68 milliseconds, an increase of 67.6%. Under extreme environmental conditions (high temperature and humidity, strong light changes), the system stability of the traditional method was 72.5%, while this method reached 95.8%, an increase of 23.3 percentage points. The system adaptability test showed that when the environmental conditions suddenly changed, the traditional method required 85 seconds to complete the adaptation adjustment, while this method only took 23 seconds, and the adjustment accuracy was improved by 58.4%.

[0073] Energy efficiency increased by 24.7%, and equipment lifespan is expected to be extended by 35.2%. User experience testing showed that lighting comfort ratings increased from 7.2 to 9.1 (out of 10), sound clarity ratings increased from 7.6 to 9.3, and overall satisfaction increased from 7.4 to 9.2. An operating cost analysis showed that this approach can save 37.8% in equipment maintenance costs, 25.6% in energy costs, and a 31.5% reduction in total cost of ownership annually.

[0074] In an optional embodiment, mapping the control scheme components using a family of hash functions, inputting the mapping results into a neural symbolic reasoning system, and generating an optimized verification sequence based on the neural symbolic reasoning system in combination with an expert knowledge base includes: Decomposing a control scheme into components to obtain a plurality of control components, mapping the plurality of control components using a hash function family to obtain initial mapping features, constructing a component feature matrix based on the initial mapping features, determining a degree of correlation between components through similarity calculation, and adjusting the component feature matrix according to the degree of correlation to generate an enhanced mapping feature; Inputting the enhanced mapping features into a neural symbolic reasoning system, extracting information from the enhanced mapping features through the neural symbolic reasoning system to construct a feature priority sequence, reconstructing the enhanced mapping features based on the feature priority sequence to obtain reconstructed features, and analyzing the temporal variation patterns of the reconstructed features to generate temporal correlation features; Construct a knowledge fusion module, establish a mapping relationship between the time series correlation features and the expert knowledge base based on the knowledge fusion module, construct a feature expression space according to the mapping relationship, convert the time series correlation features into a standard feature representation in the feature expression space, and adjust the standard feature representation based on the balance factor to generate an optimized verification sequence.

[0075] like Figure 3 As shown, the method includes: When decomposing a control scheme into its components, eigenvector decomposition techniques are used to break the original control scheme into multiple functionally independent control components. A control scheme typically contains multi-dimensional command information. For example, in a solar-powered acoustic and optical control system, a complete control scheme might include power adjustment commands, lighting angle control commands, and light intensity adjustment commands. In practical applications, a typical control scheme might contain 128 dimensions of control data, which are decomposed into eight control components, each responsible for a specific functional domain. For example, the first control component contains 16 dimensions of data and is responsible for power adjustment; the second control component contains 24 dimensions of data and is responsible for lighting angle control; and so on, forming a complete set of control components.

[0076] The hash function family mapping process uses a multi-hashing technique to assign a specific hash function to each control component. In this implementation, a hash function family is composed of multiple hash algorithms, including SHA-256, MD5, and FNV-1a. For example, the first control component has 16-dimensional raw data: {42.5, 38.7, 56.2, 44.9, 51.3, 39.8, 47.6, 53.2, 40.5, 49.1, 45.7, 52.8, 38.4, 46.2, 50.9, 43.6}. Applying the SHA-256 hash function yields a 32-bit hash value, "a7c5f3e1b8d2a4c6," which serves as the initial mapping feature for this control component. The corresponding hash functions are applied to all eight control components, generating eight sets of initial mapping features.

[0077] During the component feature matrix construction process, the eight sets of initial mapping features were arranged in rows to form an 8×32 two-dimensional matrix. Each row represents the feature representation of a control component, and each column represents the eigenvalue at a specific location. Similarity was calculated using the cosine similarity method, calculating the similarity between any two rows in the matrix. The actual calculation results show that the similarity between the first and third control components is 0.82, indicating a high correlation between the two components; the similarity between the second and fifth control components is 0.18, indicating a low correlation. Based on the calculated similarity results, feature enhancement processing was performed on component pairs with similarities greater than 0.7, amplifying their eigenvalues ​​by 20%. Component pairs with similarities less than 0.3 maintained their original values. For component pairs with similarities between 0.3 and 0.7, the eigenvalue amplification ratio was linearly proportional to the similarity. In the resulting enhanced mapping feature matrix, the eigenvalues ​​of the first and third control components were each enhanced to 1.2 times their original values, resulting in a more prominent feature expression.

[0078] The neural symbolic reasoning system employs a dual-channel architecture, consisting of a neural network processing channel and a symbolic logic reasoning channel. The neural network channel uses a bidirectional long short-term memory network structure with three hidden layers, each containing 128 neurons. The input layer receives the augmented map feature matrix and extracts high-dimensional feature representations through forward propagation. The symbolic logic reasoning channel relies on a predefined set of inference rules, including 120 if-then rules. For example, the rule "if power component value > threshold AND lighting angle component value < threshold then priority = high" determines feature priority. After the system inputs the augmented map features, the neural network channel outputs feature importance scores ranging from 0 to 1. The first control component scores 0.92, the second 0.65, the third 0.87, and so on. Based on these scores and the predefined rules, the symbolic reasoning channel generates a feature priority sequence {1, 3, 5, 2, 4, 6, 8, 7}, indicating that the first control component has the highest priority and the seventh has the lowest.

[0079] The feature reconstruction process weights the enhanced mapping features based on a priority order, with higher-priority features receiving greater weight. Specifically, the weight distribution is as follows: priority 1 components are assigned a weight of 0.35, priority 2 components are assigned a weight of 0.25, priority 3 components are assigned a weight of 0.15, and the remaining components are assigned weights of 0.1, 0.06, 0.04, 0.03, and 0.02, respectively. The reconstructed feature dimensions remain unchanged, but the values ​​of each dimension are adjusted to reflect the varying importance of different control components. A sliding window technique is used to analyze temporal variation patterns, with a window size of 10 time units and a step size of 2 time units. The feature change rate and fluctuation trend are calculated within each window. Measured data show that the change rate of the reconstructed features over 30 consecutive time units is {0.03, 0.05, 0.02, 0.07, 0.04, 0.06, 0.03, 0.08, 0.05, 0.04}, with an average change rate of 0.047. Based on this change rate sequence, time series correlation features are generated, which include three parts of information: original feature value, change rate and change trend.

[0080] The knowledge fusion module consists of an expert knowledge base and a mapping engine. The expert knowledge base contains 600 expert experience rules for solar acoustic and optical equipment, covering normal operating parameter ranges, typical failure modes, and optimized control strategies. For example, the rule "A power fluctuation rate greater than 0.1 and lasting for more than 5 minutes indicates a heat dissipation problem" is used for anomaly detection. The mapping engine uses semantic mapping technology to map time-series correlation features to knowledge base rules. In the specific implementation, the mapping similarity threshold is set to 0.75; any rule with a similarity exceeding this threshold is considered a valid mapping. For the time-series correlation features in this example, the mapping engine matched 37 relevant rules from the knowledge base, with similarities ranging from 0.76 to 0.93.

[0081] The feature representation space is constructed using a multidimensional vector space model, with eight dimensions, the same as the number of control components. In this space, each dimension represents a control capability, and the dimension values ​​are normalized to a range of 0 to 1. When converting time-series correlation features to a standard feature representation, normalization is used to map each component's value to a standard range. For example, the power control component originally ranges from 0 to 500 watts. When converted to a standard feature representation, it is mapped to a range of 0 to 1. After this conversion, the resulting eight-dimensional standard feature representation vector is {0.92, 0.65, 0.87, 0.54, 0.78, 0.43, 0.31, 0.59}.

[0082] The balance factor adjustment process introduces a dynamic balancing mechanism, fine-tuning the standard feature representation based on the overall system stability requirements. The balance factor is calculated based on the current system state and historical operating data, with a dynamic range of 0.8 to 1.2. In practice, if the system is becoming unstable, the balance factor of high-priority features is set to a value less than 1 to reduce their impact; conversely, it is set to a value greater than 1 to increase their impact. If the system load is detected to be excessive, the balance factor of the power control component is adjusted to 0.85 to reduce power output, and the balance factor of the lighting angle control component is adjusted to 1.15 to optimize the lighting effect. The adjusted standard feature representation becomes {0.78, 0.65, 0.87, 0.54, 0.90, 0.43, 0.31, 0.59}, reflecting the system's dynamic balance requirements for power and lighting control.

[0083] The resulting optimized verification sequence is a 256-bit binary sequence, with each 32 bits corresponding to the verification information for a control component. The first 32 bits of the verification sequence are "10110101110001001011010111000100," corresponding to the power control component; the next 32 bits are "01100111100100100110011110010010," corresponding to the lighting angle control component; and so on, forming the complete verification sequence. This verification sequence is transmitted along with the control plan to the execution device, which uses the same hash function family and verification algorithm to verify the integrity and validity of the control plan.

[0084] In actual application tests, the optimized verification sequence generated by this method effectively improved the system's anti-interference ability. Test data showed that when the external interference intensity was three times the standard value, the error rate of the traditional verification method was 12.7%, while the error rate of this method was only 1.3%; the verification speed increased by 42%, and the system resource usage decreased by 37%. The generation process of the optimized verification sequence took an average of 53 milliseconds, meeting the performance requirements of the real-time control system. In the stability test of 30 consecutive days of operation, the system successfully intercepted 98.7% of abnormal control instructions, effectively preventing potential equipment damage risks. Based on measured data, the application of this method in a group of solar sound and light equipment has increased the equipment's operating efficiency by 18.5%, increased energy utilization by 22.3%, and reduced maintenance costs by 35.2%, significantly enhancing the system's intelligence level and reliability.

[0085] In an optional embodiment, based on the control sequence formed by the final control scheme, the intelligent control status of the solar acoustic and optical device group is analyzed in real time to construct cross-domain collaborative data; and the generation of security data by hash mapping the cross-domain collaborative data according to the Bloom filter includes: Performing chaotic mapping on the control sequence to obtain an iterative value of the device state, constructing a chaotic mapping model based on the iterative value of the device state, analyzing the dynamic change law of the device state parameter based on the chaotic mapping model, and generating a parameter mapping sequence based on the dynamic change law; Analyzing a state change trend of a device group according to the parameter mapping sequence, wherein the state change trend reflects a co-evolution process of the device group under the action of the parameter mapping sequence, determining the operational stability of the device group based on the co-evolution process, and dynamically adjusting a control strategy according to the operational stability using the chaotic mapping model; A hash verification space is constructed for the control strategy through the chaotic mapping model, a verification path of the control strategy is established in the hash verification space, and security verification information of the control strategy is transmitted along the verification path to generate security data.

[0086] Based on the control sequence generated by the final control solution, the intelligent control status of a solar-powered acousto-optical device group is analyzed in real time. Chaotic mapping is used to process the control sequence, achieving highly secure data exchange. The control sequence contains multi-dimensional data such as power adjustment instructions, direction control instructions, and lighting intensity parameters for the acousto-optical devices. In practical applications, the control sequence can be represented as a 128-bit binary data stream, with each 8 bits representing a control instruction. When applying chaotic mapping to this control sequence, a logistic mapping function is used to process each control instruction, and the device state iteration value is iteratively calculated. Taking the power adjustment instruction as an example, with the initial power set to 380 watts, the state iteration value sequence obtained after 10 iterations is: 0.843, 0.567, 0.732, 0.615, 0.689, 0.742, 0.624, 0.578, 0.834, 0.612.

[0087] When constructing the chaotic mapping model, the aforementioned iteration value is used as the model input parameter, and a state space distribution diagram is designed, with the mapping interval being [0, 1]. In the actual device group, there are 32 solar sound and light devices, distributed across four areas, with eight devices in each area. Each device has a unique device ID, associated with a specific area in the state space. Based on this model, analysis reveals the dynamic variation pattern of the power parameter: when the iteration value is in the range [0.6, 0.8], the power regulation amplitude is relatively stable; when the iteration value is less than 0.6 or greater than 0.8, the power fluctuation increases significantly. Therefore, the stable operating parameter interval is set to [0.6, 0.8], and the abnormal fluctuation interval is set to [0, 0.6) and (0.8, 1).

[0088] The parameter mapping sequence is generated based on dynamic changes, mapping the iterative values ​​of the device state to actual physical parameters. Taking the lighting intensity parameter as an example, the mapping relationship is: lighting intensity = base lighting value + iterative value × adjustment coefficient. The base lighting value is set to 500 lumens, and the adjustment coefficient is 300 lumens. For this iterative value sequence, the generated lighting intensity parameter mapping sequence is: 752.9, 670.1, 719.6, 684.5, 706.7, 722.6, 687.2, 673.4, 750.2, 683.6 lumens. Similarly, the direction control parameter is mapped to an angle value, and the power adjustment parameter is mapped to the actual output power. The complete parameter mapping sequence contains all the control parameters for all 32 devices, forming a multidimensional control space with approximately 768 data points.

[0089] When analyzing device group status trends, the parameter mapping sequence is segmented and analyzed along the time dimension. In actual application scenarios, a time window is set to 5 minutes, and data is collected every 10 seconds, resulting in status data for 30 time points. Trend analysis of these 30 time points identifies the co-evolutionary patterns of the device group. This co-evolutionary process manifests itself as follows: when the lighting intensity of devices in a certain area increases, the lighting intensity of devices in adjacent areas decreases compensatorily; power consumption tends to be evenly distributed across the entire device group. Based on measured data, a device group is considered to be in a high stability state when the power distribution standard deviation is less than 50 watts and the lighting intensity change rate is less than 10%. A device group is considered to be in a moderate stability state when the power distribution standard deviation is between 50 and 100 watts or the lighting intensity change rate is between 10% and 20%. A device group is considered to be in a low stability state when the power distribution standard deviation is greater than 100 watts or the lighting intensity change rate is greater than 20%.

[0090] The dynamic adjustment control strategy uses an adaptive approach based on stability assessment. When a device group is detected as being in a high stability state, the current control parameters are maintained. When it is in a medium stability state, abnormal parameters are fine-tuned by 5%-10% of the current value. When it is in a low stability state, a protection mechanism is activated, adjusting the parameters of abnormal devices to a safe threshold range by up to 15%-30% of the current value. In one actual operation, for example, the standard deviation of power fluctuation of devices in Zone 2 was detected to have reached 120 watts, indicating a low stability state. The system automatically reduced the power limit of the devices in this zone by 25%, from 500 watts to 375 watts, and simultaneously adjusted the lighting angle by 15 degrees, successfully restoring system stability.

[0091] When constructing the hash verification space, the sensitivity characteristics of the chaotic map model are utilized to hash the adjusted control strategy. The dimensionality of the hash verification space corresponds to the number of devices, which is 32. The hash value for each dimension is calculated by combining the device ID and control parameters. For example, the first device, with the device ID "SL001" and control parameters including power of 350 watts, lighting angle of 42 degrees, and lighting intensity of 680 lumens, is hashed to produce the hash value "7a9c2e5f8b3d1a6". Hash values ​​are calculated for all 32 devices to form a complete hash verification space.

[0092] The verification path is established using a multi-hop verification mechanism. The verification path is set to 8 hops long, starting from the core node device and selecting the next hop device according to a preset algorithm. An example verification path is: SL001--SL008--SL015--SL022--SL004--SL026--SL019--SL032. Each hop receives the verification information from the previous hop, adds its own status information, and then passes it to the next hop. The initial verification information sent by the first hop, SL001, is 64 bytes long and includes a device status code and a timestamp. The second hop, SL008, receives the information and adds its own information, bringing the verification information to 96 bytes. This continues in this order, ultimately reaching a total of 256 bytes.

[0093] The generation of secure data is the final step after the verification path is complete. The last-hop device, the SL032, encrypts the complete verification information using a chaotic mapping model, generating a 512-byte secure data packet. This packet contains verification path information, device status, a timestamp, and an integrity check code. The characteristic of secure data is that even a single bit of information tampering will cause verification to fail. In actual testing, the security of 10,000 data transmissions was tested, successfully intercepting all simulated attack attempts, achieving a security level of 99.997%.

[0094] Verification in real-world applications has shown that the implementation of this chaotic mapping safety mechanism has extended the stable operation time of solar-powered acoustic and optical equipment clusters in harsh environments by 42%, increased power efficiency by 18%, and significantly enhanced their ability to withstand external interference. The system's adaptation time to sudden environmental changes has been shortened from 85 seconds to 31 seconds, and dynamic adjustment accuracy has been increased by 2.6 times. Data from a single year of operation in a deployed cluster of devices shows a 63% reduction in equipment failure rate, a 47% decrease in maintenance costs, and a 38% increase in user satisfaction.

[0095] A second aspect of an embodiment of the present invention provides a solar-powered acousto-optical device group collaborative control optimization system based on the Internet of Things, comprising: The first unit is used to map the position source data of the solar sound and light device group into multi-dimensional time series features to construct a device data sequence; use a long short-term memory network to analyze the feature correlation in the device data sequence, quantify the historical operation characteristics of the device, and mark the device feature identifier; The second unit is configured to calculate a device trust value based on a temporal variation trend of the device feature identifier, and divide the solar acoustic and optical device group into a core node layer and a collaborative node layer according to a distribution characteristic of the device trust value; The third unit is configured to decouple the control instruction sequence of the core node layer based on the long short-term memory network, establish a mapping relationship between the device feature identifier and the control instruction, and generate an initial control solution; establish a verification sequence for the initial control solution based on the Bloom filter of the collaborative node layer, match the device feature identifier to update the device trust value, and form a final control solution; The fourth unit is used to analyze the intelligent control status of the solar acoustic and optical equipment group in real time based on the control sequence formed by the final control scheme, and construct cross-domain collaborative data; establish a hash mapping for the cross-domain collaborative data according to the Bloom filter to generate security data, use the long short-term memory network to analyze the temporal characteristics of the security data and establish an association model with the device feature identifier to form a cross-domain resource allocation plan, and the cross-domain resource allocation plan is converted into device execution instructions through control sequence mapping.

[0096] According to a third aspect of an embodiment of the present invention, an electronic device is provided, including: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0097] According to a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described above is implemented.

[0098] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A collaborative control optimization method for solar sound and light equipment groups based on the Internet of Things, characterized in that: include: Map the position source data of the solar sound and light device group into multi-dimensional time series features to construct the device data sequence; Utilizing a long short-term memory network to analyze the feature correlation in the device data sequence, quantify the historical operation features of the device, and mark the device feature identifier; Calculating the device trust value based on the temporal change trend of the device feature identifier, and dividing the solar acoustic and optical device group into a core node layer and a collaborative node layer according to the distribution characteristics of the device trust value; Decoupling the control instruction sequence of the core node layer based on the long short-term memory network, establishing a mapping relationship between the device feature identifier and the control instruction, and generating an initial control solution; establishing a verification sequence for the initial control solution based on the Bloom filter of the collaborative node layer, matching the device feature identifier to update the device trust value, and forming a final control solution; Based on the control sequence formed by the final control scheme, the intelligent control status of the solar acoustic and optical equipment group is analyzed in real time to construct cross-domain collaborative data; according to the Bloom filter, a hash mapping is established for the cross-domain collaborative data to generate security data, and the long short-term memory network is used to analyze the temporal characteristics of the security data and establish an association model with the device feature identifier to form a cross-domain resource allocation scheme, which is converted into device execution instructions via control sequence mapping.

2. The method according to claim 1, characterized in that The position source data of the solar sound and light equipment group is mapped into multi-dimensional time series features to construct an equipment data sequence. The feature correlation in the equipment data sequence is analyzed using a long short-term memory network to quantify the historical operation characteristics of the equipment. The equipment feature identifiers include: Constructing the state source data of the solar acousto-optic device group into a time series data sequence, constructing a quantum well-quantum dot composite structure, regulating the carrier transmission path based on the quantum confinement effect of the quantum well-quantum dot composite structure, mapping the time series data sequence into a quantum state, and adjusting the energy level distribution of the quantum well-quantum dot composite structure through quantum state engineering to form a quantum feature sequence; Utilizing the quantum tunneling effect to construct a quantum transmission channel, the quantum transmission channel includes a quantum tunneling barrier array, regulating carrier transport through the Coulomb blockade effect of the quantum tunneling barrier array, and generating a device data sequence based on the coupling of the quantum tunneling effect and the quantum confinement effect; A long short-term memory network is established to parse the device data sequence, the output characteristics of the long short-term memory network are correlated and mapped with the coherence characteristics of the quantum state, the historical operation characteristics of the device are quantified based on the coherence function of the quantum state, and a device feature identifier is generated.

3. The method according to claim 2, characterized in that Constructing a quantum transmission channel using the quantum tunneling effect, the quantum transmission channel includes a quantum tunneling barrier array, regulating carrier transport through the Coulomb blockade effect of the quantum tunneling barrier array, and generating a device data sequence based on the coupling of the quantum tunneling effect and the quantum confinement effect, including: Constructing a periodically arranged quantum tunneling barrier array, establishing a carrier wave function in the barrier region of the quantum tunneling barrier array using a gradient micro-nanostructure, and modulating an exponential wave vector parameter of the carrier wave function based on the gradient micro-nanostructure; Under the action of the carrier wave function, a Coulomb blockade effect control mechanism is constructed, surface plasmon resonance is used to enhance the coupling effect of the left and right potential barriers to generate a transmission coefficient, and the tunneling current corresponding to the carrier wave function is controlled based on the transmission coefficient; The Coulomb blockade effect control mechanism and the quantum confinement effect form a coupling system, and external field control is used to realize coupled state transmission and uncoupled state transmission of the coupling system, thereby generating a coupling enhancement factor. The tunneling current is modulated according to the coupling enhancement factor to generate a device data sequence.

4. The method according to claim 1, wherein Decoupling the control instruction sequence of the core node layer based on the long short-term memory network, establishing a mapping relationship between the device feature identifier and the control instruction, and generating an initial control plan include: A long short-term memory network is trained to establish a temporal association model of device features, and the control instruction sequence of the core node layer is decoupled based on the temporal association model. A gradient feedback adjustment mechanism is used to construct a mapping relationship between the device feature identifier and the control instruction. The gradient feedback adjustment mechanism adjusts the mapping weights in real time, and an initial control scheme is generated based on the iterative optimization results of the mapping weights.

5. The method according to claim 1, wherein Establishing a verification sequence for the initial control solution according to the Bloom filter of the collaboration node layer, matching the device feature identifier to update the device trust value, and forming a final control solution includes: Constructing a verification sequence for an initial control scheme based on a Bloom filter at the collaborative node layer, decomposing the initial control scheme into multiple control scheme components, mapping the control scheme components using a family of hash functions, inputting the mapping results into a neural symbolic reasoning system, and generating an optimized verification sequence based on the neural symbolic reasoning system combined with an expert knowledge base; constructing a causal inference-driven feature matching network based on the optimized verification sequence, performing causal correlation analysis between feature components and the optimized verification sequence based on the causal inference-driven feature matching network, and calculating the device trust value using a multi-agent collaborative evaluation mechanism; The device trust value is input into a transfer learning model, and the transfer learning model generates an optimized control strategy based on the basic control strategy and the new scenario characteristics. The optimized control strategy is trusted according to the device trust value to generate a final control solution.

6. The method according to claim 5, characterized in that Mapping the control scheme components using a family of hash functions, inputting the mapping results into a neural symbolic reasoning system, and generating an optimized verification sequence based on the neural symbolic reasoning system combined with an expert knowledge base includes: Decomposing a control scheme into components to obtain a plurality of control components, mapping the plurality of control components using a hash function family to obtain initial mapping features, constructing a component feature matrix based on the initial mapping features, determining a degree of correlation between components through similarity calculation, and adjusting the component feature matrix according to the degree of correlation to generate an enhanced mapping feature; Inputting the enhanced mapping features into a neural symbolic reasoning system, extracting information from the enhanced mapping features through the neural symbolic reasoning system to construct a feature priority sequence, reconstructing the enhanced mapping features based on the feature priority sequence to obtain reconstructed features, and analyzing the temporal variation patterns of the reconstructed features to generate temporal correlation features; Construct a knowledge fusion module, establish a mapping relationship between the time series correlation features and the expert knowledge base based on the knowledge fusion module, construct a feature expression space according to the mapping relationship, convert the time series correlation features into a standard feature representation in the feature expression space, and adjust the standard feature representation based on the balance factor to generate an optimized verification sequence.

7. The method according to claim 1, characterized in that Based on the control sequence formed by the final control scheme, the intelligent control status of the solar sound and light equipment group is analyzed in real time to build cross-domain collaborative data; Establishing a hash map for the cross-domain collaborative data according to the Bloom filter to generate security data includes: Performing chaotic mapping on the control sequence to obtain an iterative value of the device state, constructing a chaotic mapping model based on the iterative value of the device state, analyzing the dynamic change law of the device state parameter based on the chaotic mapping model, and generating a parameter mapping sequence based on the dynamic change law; Analyzing a state change trend of a device group according to the parameter mapping sequence, wherein the state change trend reflects a co-evolution process of the device group under the action of the parameter mapping sequence, determining the operational stability of the device group based on the co-evolution process, and dynamically adjusting a control strategy according to the operational stability using the chaotic mapping model; A hash verification space is constructed for the control strategy through the chaotic mapping model, a verification path of the control strategy is established in the hash verification space, and security verification information of the control strategy is transmitted along the verification path to generate security data.

8. A solar-powered acoustic-optical device group collaborative control optimization system based on the Internet of Things, used to implement the method according to any one of claims 1 to 7, characterized in that: include: The first unit is used to map the position source data of the solar sound and light device group into multi-dimensional time series features to construct a device data sequence; Utilizing a long short-term memory network to analyze the feature correlation in the device data sequence, quantify the historical operation features of the device, and mark the device feature identifier; The second unit is configured to calculate a device trust value based on a temporal variation trend of the device feature identifier, and divide the solar acoustic and optical device group into a core node layer and a collaborative node layer according to a distribution characteristic of the device trust value; The third unit is configured to decouple the control instruction sequence of the core node layer based on the long short-term memory network, establish a mapping relationship between the device feature identifier and the control instruction, and generate an initial control solution; establish a verification sequence for the initial control solution based on the Bloom filter of the collaborative node layer, match the device feature identifier to update the device trust value, and form a final control solution; The fourth unit is used to analyze the intelligent control status of the solar acoustic and optical equipment group in real time based on the control sequence formed by the final control scheme, and construct cross-domain collaborative data; establish a hash mapping for the cross-domain collaborative data according to the Bloom filter to generate security data, use the long short-term memory network to analyze the temporal characteristics of the security data and establish an association model with the device feature identifier to form a cross-domain resource allocation plan, and the cross-domain resource allocation plan is converted into device execution instructions through control sequence mapping.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images