Wind Turbine Energy-Saving Intelligent Control System Based on Equipment Detection

Abstract

This invention relates to the field of building environment control and intelligent control technology for HVAC systems, and particularly to a fan energy-saving intelligent control system based on equipment detection. The system includes a condition sensing and acquisition module for acquiring the heat signal of cooking fumes, the continuous micro-pressure difference signal across the exhaust filter, the operating current signal of the exhaust fan, and the temperature and humidity sensing signals from indoor and outdoor air vents, generating a raw physical and environmental dataset. In this invention, the return air temperature and enthalpy parameters of the central air conditioning system are synchronously acquired through the IVCS host. The linkage logic is expanded from a single airflow balance to a comprehensive control mechanism that includes heat balance. While maintaining a slightly negative pressure state, cooling compensation calculations are performed based on the return air enthalpy deviation, creating a thermal-humidity coupling coordination relationship between exhaust, fresh air, and cooling adjustment, thereby improving the matching accuracy and adjustment rationality of the overall air conditioning system.

Description

Technical Field

[0001] This invention relates to the field of building environment control and intelligent control technology of HVAC, and particularly to an energy-saving intelligent control system for fans based on equipment detection. Background Technology

[0002] With the increasing complexity of modern building air conditioning systems and the rising demands for energy conservation, traditional air conditioning control logic mainly focuses on airflow balance, i.e., maintaining a stable indoor environment by adjusting fan speed and airflow. However, this control method fails to adequately consider the thermal changes in indoor temperature and humidity, especially the enthalpy changes in return air, resulting in low system regulation accuracy and low energy efficiency.

[0003] Most existing air conditioning systems rely solely on airflow regulation to maintain indoor environmental balance, neglecting the impact of heat balance. This results in the air conditioning system's response being inflexible and unable to accurately adjust the temperature and humidity when the outdoor environment or load changes, leading to insufficient energy efficiency and reduced indoor comfort. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides a fan energy-saving intelligent control system based on equipment detection, which aims to improve the problem that most existing air conditioning systems rely solely on air volume regulation to maintain the balance of the indoor environment, ignoring the impact of heat balance.

[0005] This invention provides the following technical solution: a wind turbine energy-saving intelligent control system based on equipment detection, comprising: The working condition sensing and acquisition module is used to acquire the oil fume heat signal in the cooking area, the continuous micro pressure difference signal on both sides of the exhaust filter, the operating current signal of the exhaust fan, and the temperature and humidity sensing signals of the indoor and outdoor air outlets, and generate the original physical and environmental dataset. The resistance state extraction constraint module, based on the original physical and environmental dataset, analyzes the spatiotemporal coupling relationship between the continuous micro-pressure difference signal and the operating current signal of the exhaust fan through a fluid dynamics model, extracts the pipeline resistance index, and aligns and compares the pipeline resistance index with a preset baseline to generate a pipeline frequency limiting constraint matrix. The environmental enthalpy calculation and analysis module, based on the temperature and humidity sensing signals of the indoor and outdoor air outlets in the original physical and environmental dataset, uses the enthalpy-humidity map algorithm model to perform thermodynamic calculations and outputs indoor and outdoor environmental enthalpy vectors. The feature fusion coupling module receives the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector, calculates the correlation factor between the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector through feature splicing and weight verification, and generates a global ratio descriptor. The frequency conversion decision evaluation module calculates the target exhaust baseline volume based on the global ratio descriptor and the oil fume thermal signal in the original physical and environmental dataset, and generates a speed regulation decision parameter set through multivariate nonlinear solution. The control signal generation module generates exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands respectively based on the speed regulation decision parameter set and through preset mapping rules, and integrates and encapsulates them into a control command cluster. The execution module drives the associated exhaust fan, fresh air fan, and central air conditioning system to perform corresponding frequency conversion and cooling capacity adjustment operations based on the control command set.

[0006] Preferably, in the condition sensing and acquisition module, generating the original physical and environmental dataset specifically includes the following steps: The cooking area's oil fume heat signal, the continuous micro-pressure difference signal on both sides of the exhaust filter, the exhaust fan's operating current signal, and the indoor and outdoor air vents' temperature and humidity sensing signals are collected by a preset sampling frequency, and then processed by analog-to-digital conversion to generate a discrete digital signal sequence. The arrival timestamps of the discrete digital signal sequence are extracted, the relative time offset is calculated based on the system global clock, and the discrete digital signal sequence is time-aligned by an interpolation resampling algorithm to generate a synchronization multidimensional state matrix. The synchronous multidimensional state matrix is ​​input into a preset filtering model, and the high-frequency environmental noise components in the synchronous multidimensional state matrix are separated and removed to output the filtering feature matrix. Extract each data dimension vector from the filtered feature matrix, and according to a preset data encapsulation protocol, structurally concatenate each data dimension vector along with the corresponding timestamp index and sensor node identifier to generate the original physical and environmental dataset.

[0007] Preferably, in the resistance state extraction constraint module, the generation of the pipeline network frequency limiting constraint matrix specifically includes the following steps: The continuous micro-pressure difference signal and the operating current signal of the exhaust fan are extracted from the original physical and environmental dataset, and the features are spliced ​​according to the time series dimension to generate resistance feature vector pairs. The resistance feature vector is input into a preset fluid dynamics model, and the spatiotemporal coupling coefficient between the continuous micro-pressure difference signal and the operating current signal of the exhaust fan is calculated through polynomial fitting, and the pipeline resistance index is output. Extract the pipeline resistance index, align the pipeline resistance index with a preset baseline and perform differential operation to generate a resistance attenuation deviation that characterizes the current pipeline blockage state; The resistance attenuation deviation is input into a preset frequency limiting mapping function to calculate the upper limit threshold of the frequency constraint. The upper limit threshold of the frequency constraint is then reconstructed into a matrix according to the system control dimension to generate the pipeline frequency limiting constraint matrix.

[0008] Preferably, in the environmental enthalpy calculation and analysis module, the output of indoor and outdoor environmental enthalpy vectors specifically includes the following steps: The temperature and humidity sensing signals of the indoor and outdoor air outlets are extracted from the original physical and environmental dataset and classified and packaged according to the spatial identifiers of indoor and outdoor spaces to generate a set of environmental thermal parameters. The environmental thermal parameter component set is input into the enthalpy-humidity diagram algorithm model. The specific enthalpy values ​​of indoor and outdoor measuring points at the current sampling time are obtained by solving the preset air state equation, and the instantaneous specific enthalpy data pairs are output. Perform time-domain filtering based on a sliding window on the instantaneous enthalpy data pairs, calculate the statistical mean of enthalpy values ​​within a preset sampling period, and generate a mean enthalpy value sequence; According to the preset vector construction rules, the mean enthalpy sequence is mapped to the corresponding multidimensional spatial coordinate system to generate indoor and outdoor environmental enthalpy vectors.

[0009] Preferably, in the feature fusion coupling module, the generation of the global ratio descriptor specifically includes the following steps: Obtain the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vectors, perform data dimension alignment and feature splicing operations, and generate a cross-domain feature fusion matrix; The cross-domain feature fusion matrix is ​​input into a preset weight allocation model to perform weight verification operations and output a weighted feature sequence. Perform covariance analysis on the weighted feature sequence to calculate and extract the numerical correlation factor between the pipeline frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector; According to the preset format encapsulation protocol, the numerical correlation factors are subjected to multidimensional coordinate mapping and structured recombination to generate a global ratio descriptor.

[0010] Preferably, in the frequency conversion decision evaluation module, the generation of the speed regulation decision parameter set specifically includes the following steps: The oil fume thermal signal is extracted from the original physical and environmental dataset, and the oil fume thermal signal is input into a preset load mapping function for numerical conversion to calculate the target exhaust baseline volume. Obtain the global ratio descriptor, align the target exhaust baseline quantity with the global ratio descriptor using data dimension alignment and multi-dimensional feature fusion, and generate the working condition state vector to be solved. The state vector of the working condition to be solved is input into a preset multivariable nonlinear equation system to perform iterative analytical calculations and output the target frequency values ​​of each control node. According to the preset hardware control protocol, the target frequency value is structured and encapsulated to generate a speed regulation decision parameter set.

[0011] Preferably, in the control signal generation module, the fusion and encapsulation into a control instruction cluster specifically includes the following steps: Extract the numerical components of each control node in the speed regulation decision parameter set, and classify them according to the preset hardware device address table to generate a classification parameter matrix. The classification parameter matrix is ​​input into the preset mapping rule to perform control protocol conversion, generating exhaust frequency conversion command, fresh air linkage command and air conditioning adjustment command respectively; Extract the exhaust frequency conversion command, the fresh air linkage command, and the air conditioning adjustment command, and add a system global clock stamp to the exhaust frequency conversion command, the fresh air linkage command, and the air conditioning adjustment command, and perform timing alignment operation to generate a timing alignment command set; According to the preset underlying bus communication data frame format, the timing alignment instruction set is processed by message packing and checksum filling operations to generate a control instruction set.

[0012] Preferably, in the execution module, the corresponding frequency conversion and cooling capacity adjustment operations performed by the drive-associated exhaust fan, fresh air unit, and central air conditioning system specifically include the following steps: The control instruction set is disassembled by the underlying protocol parser to extract the exhaust frequency conversion instruction, the fresh air linkage instruction, and the air conditioning adjustment instruction; Establish a logical mapping relationship between the exhaust frequency conversion command, the fresh air linkage command, and the air conditioning adjustment command and the preset physical communication port, and generate a sequence of commands to be distributed; According to the encoding format of the exhaust frequency conversion instruction, the fresh air linkage instruction, and the air conditioning adjustment instruction in the instruction sequence to be distributed, the exhaust frequency conversion instruction, the fresh air linkage instruction, and the air conditioning adjustment instruction are converted into underlying physical drive signals that meet the hardware communication protocol requirements. The underlying physical drive signals are output to the first frequency converter associated with the exhaust fan, the second frequency converter associated with the fresh air fan, and the cooling capacity regulating valve controller associated with the central air conditioning system, respectively, to complete the frequency conversion and cooling capacity regulation operations.

[0013] The present invention has the following beneficial effects: 1. In this invention, the return air temperature and enthalpy parameters of the central air conditioning are synchronously acquired through the IVCS host, and the linkage logic is expanded from a single air volume balance to a comprehensive control mechanism that includes heat balance. While maintaining a slightly negative pressure state, the cooling capacity compensation calculation is performed based on the return air enthalpy deviation, so that a thermal and humidity coupling coordination relationship is formed between exhaust air, fresh air and cooling capacity adjustment, thereby improving the matching accuracy and adjustment rationality of the overall air conditioning system.

[0014] 2. In this invention, by forming a standardized control instruction link through precision scaling quantization, address concatenation, protocol encapsulation, and physical port mapping of the upper-layer floating-point frequency decision results, a complete data closed loop from algorithm decision to hardware execution is established. This enables direct multi-device linkage control based on a clear data structure and mapping relationship, thereby improving the system's feasibility and engineering feasibility.

[0015] 3. In this invention, by introducing an enthalpy compensation algorithm on the basis of the original air volume collaborative control and combining it with a time synchronization execution mechanism, the control decision not only considers the air volume flow relationship but also the changes in sensible and latent heat carried by a unit mass of air, thereby enhancing the system's adaptability under seasonal changes or load fluctuations and improving operational stability and energy efficiency.

[0016] 4. By adding a filter resistance monitoring module to the system and collecting inlet and outlet air pressure difference data in real time, the filter resistance change trend can be calculated in conjunction with the wind speed, and the fan frequency can be adjusted or a maintenance warning can be issued in linkage. This allows the equipment to maintain the target air volume and reasonable energy consumption level even as the filter gradually becomes clogged. At the same time, it reduces the additional losses caused to the motor and frequency converter by long-term high-load operation, achieving a balance between energy saving and preventive maintenance. Attached Figure Description

[0017] Figure 1 This is an architecture diagram of the wind turbine energy-saving intelligent control system based on equipment detection proposed in this invention; Figure 2 This is a schematic diagram of the system composition and connection relationship proposed in this invention. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides a wind turbine energy-saving intelligent control system based on equipment detection, such as... Figure 1 and Figure 2 As shown, it includes: The working condition sensing and acquisition module is used to acquire the oil fume heat signal in the cooking area, the continuous micro pressure difference signal on both sides of the exhaust filter, the operating current signal of the exhaust fan, and the temperature and humidity sensing signals of the indoor and outdoor air outlets, and generate the original physical and environmental dataset. Furthermore, the process of generating the original physical and environmental dataset in the condition sensing and acquisition module specifically includes the following steps: The cooking area's oil fume heat signal, the continuous micro-pressure difference signal on both sides of the exhaust filter, the exhaust fan's operating current signal, and the temperature and humidity sensing signals from indoor and outdoor air vents are collected by a preset sampling frequency. These signals are then processed by analog-to-digital conversion to generate a discrete digital signal sequence. The arrival timestamps of discrete digital signal sequences are extracted, the relative time offset is calculated based on the system global clock, and the discrete digital signal sequences are time-aligned using an interpolation resampling algorithm to generate a synchronization multidimensional state matrix. The synchronous multidimensional state matrix is ​​input into the preset filtering model, the high-frequency environmental noise components in the synchronous multidimensional state matrix are separated and removed, and the filtering feature matrix is ​​output. Extract the data dimension vectors from the filter feature matrix, and according to the preset data encapsulation protocol, combine the data dimension vectors with the corresponding timestamp index and sensor node identifier in a structured manner to generate the original physical and environmental dataset.

[0020] Specifically, in the operating environment of commercial kitchen exhaust systems or industrial ventilation systems, the operating condition sensing and acquisition module first collects information from multiple types of underlying physical sensors through a preset sampling frequency. The system sets a fixed sampling period to cyclically acquire the analog voltage signal from the oil fume heat sensor in the cooking area, the continuous micro differential pressure signal fed back by the differential pressure transmitter installed on both sides of the exhaust filter, the operating current signal collected by the current transformer in the exhaust fan power supply circuit, and the temperature and humidity sensing signals distributed in the indoor and outdoor air outlets. After the above multi-dimensional analog physical signals are transmitted to the main control board, they are quantized through the built-in analog-to-digital conversion processing to generate discrete digital signal sequences corresponding to each physical acquisition channel.

[0021] Because of the differences in response delay between the various sensor hardware and the transmission rate of the underlying communication bus, the system further extracts the arrival timestamps of the discrete digital signal sequences and calculates the relative time offset based on the system's internal master global clock. Based on this, the system performs timing alignment of each discrete digital signal sequence using an interpolation resampling algorithm. The specific linear interpolation analytical algorithm is as follows: ; In the formula, Indicates the target synchronization sampling time determined by the system's global clock; and These represent the original arrival timestamps of the discrete digital signal sequences of the i-th physical dimension collected before and after the synchronous sampling time of the target, respectively. and These represent the specific physical quantity values ​​collected at the corresponding original arrival timestamp, specifically the micro-pressure difference measurement value on both sides of the exhaust filter, the ampere value of the exhaust fan operating current, or the ambient temperature and humidity values. Indicates the relative time offset, and ; This represents the synchronization feature value of the i-th dimension generated at the target synchronization sampling time after interpolation calculation.

[0022] The system performs resampling calculations on signals related to oil fume heat, micro-pressure difference, operating current, and ambient temperature and humidity. It then stacks the synchronous feature values ​​of all dimensions at the same target synchronous sampling time t according to a preset sensor array sequence, generating a synchronous multidimensional state matrix where all physical data features are completely consistent on the time axis. The system then inputs this generated synchronous multidimensional state matrix into a preset first-order low-pass digital filter model to separate and remove high-frequency environmental noise components caused by high-frequency airflow disturbances in the duct or power grid harmonics. The specific analytical algorithm for this digital filter model is as follows: ; In the formula, This represents the synchronization feature value of the i-th sensor physical dimension in the synchronous multidimensional state matrix at the k-th global synchronization sampling time. This represents the currently calculated filter characteristic value; This represents the historical filtered feature value corresponding to the physical dimension at the previous synchronous sampling moment. At the first sampling moment after system startup, i.e., when k equals 1, the preset... As initial boundary conditions; This represents the preset filter coefficient corresponding to a specific physical dimension of the sensor signal, which is specifically given by the system based on the frequency of airflow disturbance and electrical harmonic characteristics in the duct.

[0023] By performing iterative filtering calculations on various data dimensions such as differential pressure and operating current, the system finally outputs a filtering feature matrix that removes high-frequency physical noise. After completing the filtering process, the system extracts the data dimension vectors from the filtering feature matrix and strictly follows the preset underlying data encapsulation communication protocol to align the differential pressure dimension vector, current dimension vector, thermal dimension vector, and temperature and humidity dimension vector to the specified byte length. Together with the corresponding consistent timestamp index and the node identifier of each sensor on the bus, the data is structurally spliced ​​together. The spliced ​​data forms a continuous memory data block, generates the original physical and environmental dataset, and stores it in the system's shared storage area.

[0024] This solves the problems of time misalignment and high-frequency noise interference from multi-source heterogeneous sensors, ensuring the consistency of the underlying acquired data in terms of timing and values.

[0025] The resistance state extraction constraint module, based on the original physical and environmental dataset, analyzes the spatiotemporal coupling relationship between the continuous micro-pressure difference signal and the operating current signal of the exhaust fan through a fluid dynamics model, extracts the pipeline network resistance index, and aligns and compares the pipeline network resistance index with a preset baseline to generate a pipeline network frequency limiting constraint matrix. Furthermore, in the resistance state extraction constraint module, generating the pipeline network frequency limiting constraint matrix specifically includes the following steps: Continuous micro-pressure difference signals and exhaust fan operating current signals are extracted from the original physical and environmental datasets, and features are concatenated according to the time series dimension to generate resistance feature vector pairs. The resistance feature vector is input into a preset fluid dynamics model, and the spatiotemporal coupling coefficient between the continuous micro-pressure difference signal and the operating current signal of the exhaust fan is calculated through polynomial fitting, and the pipeline resistance index is output. Extract the pipeline resistance index, align the pipeline resistance index with the preset baseline and perform differential operation to generate the resistance attenuation deviation that characterizes the current pipeline blockage state; The resistance attenuation deviation is input into the preset frequency limiting mapping function to calculate the upper limit threshold of the frequency constraint. The upper limit threshold of the frequency constraint is then reconstructed into a matrix according to the system control dimension to generate the pipeline frequency limiting constraint matrix.

[0026] Specifically, after the system completes the synchronous acquisition and feature extraction of physical signals, the resistance state extraction constraint module retrieves the generated original physical and environmental dataset from the aforementioned system shared storage area. The system, according to the set sensor node identifiers, directionally extracts the filtered continuous micro-pressure difference signal and the exhaust fan's operating current signal from this original physical and environmental dataset. After extraction, the system uses the global synchronization timestamp as a reference and performs matrix column concatenation of the continuous micro-pressure difference numerical sequence and the corresponding operating current ampere value sequence along the time series dimension to generate a resistance feature vector pair characterizing the current fluid transport condition. Because the long-term adsorption of oil stains inside the commercial kitchen exhaust duct network leads to a reduction in the effective flow cross-sectional area, the aerodynamic resistance characteristics of the duct network will dynamically evolve accordingly. The system inputs the generated resistance feature vector pair into a preset fluid dynamics model, and based on the fan similarity law and the duct network characteristic curve theory, calculates the spatiotemporal coupling coefficient between the continuous micro-pressure difference signal and the exhaust fan's operating current signal through polynomial fitting. The specific least-squares polynomial fitting objective analytical algorithm is as follows: ; In the formula, This represents the total number of discrete-time sampling points contained in the currently extracted resistance feature vector pair; This represents the specific value of the continuous micro-pressure difference signal extracted at the k-th time sampling point. This value characterizes the physical quantity of static pressure drop at both ends of the exhaust filter or local duct. This represents the specific value of the operating current signal of the exhaust fan extracted at the k-th time sampling point; This represents the set of spatiotemporal coupling coefficients to be solved, containing variables. and ; Characterizes the second-order fluid retardation coefficient, reflecting the dynamic pressure loss characteristics of the fluid in the pipe network; Characterizes the first-order linear friction coefficient along the pipe wall, reflecting the viscous resistance characteristics of the pipe wall; This represents the fitting error cost function. The system solves for this using the partial derivative zeroing method, making... Get the minimum value Specific numerical values, and the solution representing the dynamic pressure loss. The numerical output is a pipeline resistance index. After obtaining the pipeline resistance index, the system retrieves the baseline resistance coefficient of the pipeline in a clean initial installation state from the local memory as the preset baseline. The system extracts the above-output pipeline resistance index, spatially aligns this pipeline resistance index with the preset baseline according to data type, and performs a difference operation. The specific difference analysis algorithm is as follows: ; In the formula, This represents the extracted pipeline resistance index; This indicates the baseline hysteresis coefficient value retrieved from the preset baseline. This represents the resistance attenuation deviation, generated through differential operations, indicating the current state of pipeline blockage. To prevent the fan from operating at full speed under severely blocked pipeline conditions, which could lead to motor overload and overheating, the system inputs the generated resistance attenuation deviation into a preset frequency limiting mapping function. The specific frequency limiting parsing algorithm is as follows: ; In the formula, This represents the drag attenuation deviation calculated using the above steps; This indicates the upper limit threshold of the rated physical operating frequency of the exhaust fan inverter; This represents the preset penalty derating factor, which is specifically obtained from the overload protection characteristic curve of the motor at the factory. This represents the calculated upper limit threshold for the frequency constraint of the exhaust fan. After obtaining this value, the system reconstructs this upper limit threshold into a matrix according to the system control dimensions such as exhaust and fresh air. The specific matrix reconstruction parsing algorithm is as follows: ; In the formula, This represents the upper limit threshold of the frequency constraint for the exhaust fan calculated above; This indicates the system's preset indoor micro-negative pressure maintenance ratio coefficient; This represents the upper limit threshold of the frequency constraint for the fresh air unit, calculated through a linkage mechanism. This represents the reconstructed network frequency limiting constraint matrix. The system stores this network frequency limiting constraint matrix in the cache space for subsequent use by the frequency conversion decision evaluation module.

[0027] By fitting the underlying fluid data, the physical state of the actual blockage in the duct was numerically quantified, and the corresponding frequency reduction boundary of the inverter was calculated based on the quantification result, thus establishing the operating limit conditions of the control process.

[0028] The environmental enthalpy calculation and analysis module uses the temperature and humidity sensor signals of indoor and outdoor air outlets in the original physical and environmental dataset to perform thermodynamic calculations using the enthalpy-humidity map algorithm model and outputs indoor and outdoor environmental enthalpy vectors. Furthermore, in the environmental enthalpy calculation and analysis module, the output of indoor and outdoor environmental enthalpy vectors specifically includes the following steps: Temperature and humidity sensing signals from indoor and outdoor air vents are extracted from the original physical and environmental dataset, and classified and encapsulated according to indoor and outdoor spatial identifiers to generate environmental thermal parameter component sets. The environmental thermal parameter component set is input into the enthalpy-humidity map algorithm model. The specific enthalpy values ​​of indoor and outdoor measuring points at the current sampling time are obtained by solving the preset air state equation, and the instantaneous specific enthalpy data pairs are output. Perform time-domain filtering based on a sliding window on the instantaneous enthalpy data pairs, calculate the statistical mean of enthalpy values ​​within a preset sampling period, and generate a mean enthalpy value sequence; According to the preset vector construction rules, the mean enthalpy value sequence is mapped to the corresponding multidimensional spatial coordinate system to generate indoor and outdoor environmental enthalpy vectors.

[0029] Specifically, after the aforementioned condition sensing and acquisition module generates and stores the original physical and environmental dataset in the shared storage area, the environmental enthalpy value calculation and analysis module retrieves the required parameters from the continuous memory data blocks of this dataset. The system locates the sensor node identifiers in the data packet and extracts the temperature and humidity sensing signals representing the locations of the indoor kitchen area air vents and outdoor air inlets. The system classifies, parses, and encapsulates the extracted temperature and humidity sensing signals according to the preset indoor and outdoor spatial physical location identifiers, treating indoor dry-bulb temperature and relative humidity as one independent subset and outdoor dry-bulb temperature and relative humidity as another independent subset, generating a set of environmental thermal parameter components for thermodynamic calculations. The system inputs the generated set of environmental thermal parameter components into the underlying enthalpy-humidity map algorithm model. This model, based on the thermodynamic properties of moist air, first calculates the current air humidity using a preset moist air state conversion formula. The specific humidity content analysis algorithm is as follows: ; In the formula, This represents the percentage of relative humidity in the set of environmental thermal parameters extracted at the current k-th sampling time. This represents the saturated water vapor partial pressure value at the same temperature, obtained by looking up a table or calculating based on the current dry-bulb temperature within the model. This represents the preset local standard atmospheric pressure constant; This represents the calculated air humidity content. The system then obtains the specific enthalpy values ​​of indoor and outdoor measuring points at the current sampling time by solving a preset air state equation. The specific instantaneous specific enthalpy calculation algorithm is as follows: ; In the formula, This represents the dry-bulb temperature in degrees Celsius extracted from the set of environmental thermal parameters at the current k-th sampling time, specifically corresponding to the extracted dry-bulb temperature of the indoor or outdoor measuring point. This represents the preset specific heat capacity constant of dry air at constant pressure; This represents the latent heat of vaporization of water at zero degrees Celsius (preset). This represents the preset isobaric specific heat capacity constant of water vapor; This represents the instantaneous enthalpy of moist air at the current sampling moment, calculated through equation solving. The system performs the aforementioned thermodynamic calculations on data from both indoor and outdoor measuring points, aligning and pairing the indoor and outdoor instantaneous enthalpy values ​​at the same target sampling moment along the time axis, outputting instantaneous enthalpy data pairs. To eliminate interference from instantaneous fluctuations in local airflow temperature and humidity caused by movement of people and frequent door openings in the kitchen area, the system performs time-domain filtering based on a sliding window on the output instantaneous enthalpy data pairs. The system calculates the statistical mean of enthalpy values ​​within a preset sampling period within a set time window. The specific sliding window mean analysis algorithm is as follows: ; In the formula, This represents the indoor instantaneous enthalpy value calculated and output at the k-th (j-th)th historical sampling time within the sliding window; This indicates the total number of discrete-time sampling points included in the preset sliding time window; This represents the indoor steady-state mean enthalpy calculated at the k-th time. The system simultaneously performs the same mean calculation process on the outdoor instantaneous enthalpy value to obtain the outdoor steady-state mean enthalpy. This indoor and outdoor steady-state mean enthalpy are then arranged in parallel according to time series, generating a mean enthalpy sequence. Subsequently, the system maps the generated mean enthalpy sequence to the corresponding multi-dimensional coordinate system for structured representation according to preset vector construction rules. The specific vector mapping construction analytical algorithm is as follows: ; In the formula, This represents the indoor steady-state mean specific enthalpy calculated above; This represents the outdoor steady-state mean specific enthalpy calculated above; This represents the numerical component of the enthalpy difference, which characterizes the difference in thermal gradient between indoors and outdoors, obtained by subtracting the average enthalpy of the indoor steady-state enthalpy from the average enthalpy of the outdoor steady-state enthalpy. This represents the reconstructed indoor and outdoor environmental enthalpy vector. The system stores this vector in a specific register of the main control chip for subsequent system use in feature fusion and air conditioning cooling capacity ratio calculations.

[0030] Through physical-thermodynamic conversion, the surface temperature and humidity sensing signals are accurately analyzed into core parameters reflecting the total thermal energy of the air, providing an objective data benchmark for subsequent load linkage decisions of the system.

[0031] The feature fusion coupling module receives the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vectors. Through feature splicing and weight verification, it calculates the correlation factor between the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vectors, and generates a global matching descriptor. Furthermore, in the feature fusion coupling module, generating the global matching descriptor specifically includes the following steps: Obtain the network frequency limiting constraint matrix and indoor and outdoor environmental enthalpy vectors, perform data dimension alignment and feature stitching operations, and generate a cross-domain feature fusion matrix; The cross-domain feature fusion matrix is ​​input into a preset weight allocation model to perform weight verification operations and output a weighted feature sequence. Perform covariance analysis on the weighted feature sequence to calculate the numerical correlation factor between the extracted pipeline frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector; According to the preset format encapsulation protocol, the numerical correlation factors are subjected to multidimensional coordinate mapping and structured recombination to generate a global ratio descriptor.

[0032] Specifically, after the aforementioned modules complete the resistance assessment and thermodynamic calculation, the feature fusion coupling module reads the previously generated pipeline frequency limiting constraint matrix and indoor / outdoor environmental enthalpy vector from the system cache space and specific registers of the main control chip, respectively. The system extracts the main diagonal elements of the pipeline frequency limiting constraint matrix, namely the upper limit threshold of the frequency constraint for the exhaust fan and the upper limit threshold of the frequency constraint for the fresh air fan, and combines them into a one-dimensional mechanical limiting vector. At the same time, the system retrieves the indoor steady-state mean ratio enthalpy and the enthalpy difference component representing the difference between the indoor and outdoor thermal gradients from the indoor / outdoor environmental enthalpy vector, and performs data dimension standardization scaling to eliminate the absolute dimensional difference between the frequency (Hertz level) and the enthalpy (kilojoule level).

[0033] After dimensional alignment, the system concatenates the one-dimensional mechanical limiting vector with the indoor and outdoor environmental enthalpy vectors at the current sampling time to generate a transient cross-domain state vector. Within a set evaluation period, the system continuously extracts transient cross-domain state vectors from multiple historical sampling times and stacks them column-wise according to their chronological order, ultimately generating a cross-domain feature fusion matrix containing both temporal and spatial two-dimensional attributes.

[0034] The system then inputs the generated cross-domain feature fusion matrix into a preset weight allocation model to perform weight verification calculations. Due to the significant difference between the mechanical response physical inertia of the fan equipment and the hysteresis of changes in the indoor air thermodynamic state, the system assigns different physical importance coefficients to the mechanical limiting feature and the environmental enthalpy feature in the aforementioned matrix. The specific weight verification parsing algorithm is as follows: ; In the formula, This represents the specific value of the i-th physical dimension in the cross-domain feature fusion matrix at the k-th historical sampling time. This represents the preset verification weight coefficient for the i-th physical dimension, specifically divided into mechanical response weight determined by the physical inertia of the system motor rotor and thermal hysteresis weight determined by the physical volume of the kitchen space. This represents the weighted feature value of the i-th physical dimension at the k-th sampling time, output after verification. The system defines i equal to 1 as the mechanical limiting physical dimension, and the corresponding feature sequence generated throughout the entire evaluation period is denoted as the weighted mechanical limiting feature sequence. The system defines i equal to 2 as the environmental enthalpy difference physical dimension, and the corresponding feature sequence generated throughout the entire evaluation period is denoted as the weighted environmental enthalpy difference feature sequence. To accurately identify the underlying mutual constraint relationship between the pipeline physical blockage limit and the air conditioning cooling capacity compensation load, the system performs covariance analysis on the aforementioned explicitly generated weighted mechanical limiting feature sequence and weighted environmental enthalpy difference feature sequence to calculate and extract the numerical correlation factor between the pipeline frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vectors. The specific covariance extraction analysis algorithm is as follows: ; In the formula, This indicates the total number of historical time sampling points included within the aforementioned evaluation period; This represents the specific weighted value of the weighted mechanical limiting feature sequence generated when i equals 1 at the j-th historical sampling time; This represents the statistical arithmetic mean of the weighted mechanical limiting feature sequence over the current N sampling point evaluation period; This represents the specific weighted value of the weighted environmental enthalpy difference feature sequence generated when i equals 2 at the j-th historical sampling time; This represents the statistical arithmetic mean of the weighted environmental enthalpy difference characteristic sequence over the current N sampling point evaluation period; This represents a numerical correlation factor, calculated using covariance, reflecting the degree of coupling between mechanical constraints and thermal load. This factor quantifies the sensitivity of indoor heat accumulation to the impact of reduced ventilation capacity due to pipe blockage. After calculating this factor, the system encapsulates the communication protocol according to a preset format and performs multi-dimensional coordinate mapping and structured reorganization on the numerical correlation factor. The specific reorganization descriptor parsing algorithm is as follows: ; In the formula, This represents the core numerical correlation factor obtained from the above solution; This represents the instantaneous weighted value of the weighted mechanically limited feature sequence at the latest sampling time with time subscript N; This represents the instantaneous weighted value of the weighted environmental enthalpy difference characteristic sequence at the latest sampling time with time subscript N; This represents the global allocation descriptor generated after multidimensional coordinate mapping and structured recombination.

[0035] The system stores the global ratio descriptor in the main control memory area and transmits it to the next-level frequency conversion decision evaluation module as the core boundary input parameter for solving the nonlinear equation.

[0036] By using rigorous mathematical algebraic mapping, heterogeneous physical quantities are transformed to a unified scale, and the degree of physical interference between the drag state and the thermodynamic state is quantified, thus establishing multidimensional composite variable boundary conditions for the subsequent nonlinear control solution of the system.

[0037] The variable frequency decision evaluation module calculates the target exhaust baseline based on the global ratio descriptor and the oil fume thermal signal in the original physical and environmental dataset, and generates a set of speed regulation decision parameters through multivariate nonlinear solution. Furthermore, in the frequency converter decision evaluation module, generating the speed control decision parameter set specifically includes the following steps: The oil fume thermal signal is extracted from the original physical and environmental dataset, and the oil fume thermal signal is input into a preset load mapping function for numerical conversion to calculate the target exhaust baseline volume. Obtain the global ratio descriptor, align the target exhaust baseline with the global ratio descriptor using data dimensions and fuse multi-dimensional features to generate the state vector of the working condition to be solved. The state vector of the working condition to be solved is input into a preset multivariable nonlinear equation system to perform iterative analytical calculations and output the target frequency values ​​of each control node. According to the preset hardware control protocol, the target frequency value is structured and encapsulated to generate a set of speed regulation decision parameters.

[0038] Specifically, after the aforementioned feature fusion coupling module generates a global ratio descriptor and stores it in the main control memory area, the frequency conversion decision evaluation module retrieves the original physical and environmental dataset from the system's shared storage area. Based on the corresponding sensor physical node identifiers, the system selectively extracts specific voltage acquisition values ​​of the oil fume thermal signal reflecting changes in the cooking area from this dataset. The system inputs the oil fume thermal signal extracted at the latest sampling time into a preset load mapping function for polynomial numerical conversion. The specific exhaust reference volume analysis algorithm is as follows: ; In the formula, This represents the specific voltage value of the oil fume thermal signal extracted from the original physical and environmental dataset; This indicates the baseline reference voltage value when the system's preset cooking zone is in a non-working, no-load state; This represents the preset quadratic nonlinear mapping coefficient that characterizes the increase in heat generation of oil fumes and the increase in fan response; This indicates the preset minimum physical operating frequency threshold for the exhaust fan to maintain the basic flow velocity of the pipeline network; This represents the target exhaust baseline volume calculated after conversion. After generating the target exhaust baseline volume, the system retrieves the generated global allocation descriptor from the autonomous control memory area.

[0039] The system extracts the numerical correlation factors encapsulated within the descriptor, the weighted instantaneous weighted value of the weighted mechanical limiting at the latest sampling moment, and the weighted instantaneous weighted value of the weighted environmental enthalpy difference. The system then performs memory alignment of the target exhaust baseline quantity with the extracted feature parameters in terms of data dimension and data type, and performs multi-dimensional feature fusion according to the set column vector arrangement order to generate a one-dimensional state vector of the operating condition to be solved. The specific state vector construction parsing algorithm is as follows: ; In the formula, This represents the target exhaust baseline volume calculated above; This represents a numerical correlation factor extracted from the global allocation descriptor, reflecting the degree of coupling between mechanical constraints and thermal loads. This represents the instantaneous weighted value of the weighted mechanically limited feature sequence extracted from the global allocation descriptor at the latest sampling moment; This represents the instantaneous weighted value of the weighted environmental enthalpy difference feature sequence extracted from the global ratio descriptor at the latest sampling time; This represents the state vector of the operating condition to be solved, generated through feature stacking and fusion. The system then inputs this state vector into a pre-defined set of multivariable nonlinear equations for iterative analytical computation. Due to the physical interference between the indoor micro-negative pressure maintenance ratio and the exhaust resistance limitation, the system employs a nonlinear Gauss-Seidel iterative method for multivariable coupled solution. The specific nonlinear iterative solution algorithm is as follows: ; ; ; In the formula, This indicates the iteration step number of the current iterative analytical operation. The system defaults to an initial iteration step number m equal to zero. This is the minimum fresh air operation frequency of the system; This represents the preset exhaust resistance coupling compensation amplification factor; This represents the preset exhaust and fresh air airflow coupling feedback compensation coefficient; This indicates the preset indoor micro-negative pressure maintenance ratio coefficient; This indicates the preset cold and heat penetration inhibition coefficient; This indicates the preset inverter conversion factor for the air conditioner compressor; This indicates that the absolute value of the instantaneous weighted value of the environmental enthalpy difference is taken to eliminate the anomaly of no solution in the logarithmic domain calculation of the opposite thermal gradients in winter and summer; This represents the target frequency value of the exhaust fan control node calculated and output in the m+1th iteration; This represents the target frequency value of the fresh air unit control node calculated and output in the m+1th iteration; This represents the target frequency value of the central air conditioning variable frequency compressor control node calculated and output in the m+1th iteration.

[0040] The system continues to perform the above iterative calculations until the absolute value of the difference between two consecutive iterations of the target frequency value of the exhaust fan is less than the preset convergence tolerance threshold. At this point, the iteration stops, and the final converged target frequency values ​​for each control node are output. After obtaining the converged solution results, the system extracts the three calculated target frequency values ​​and performs data conversion operations strictly according to the preset hardware control protocol. The system converts the decimal frequency values ​​into hexadecimal machine code that meets the requirements of the underlying bus communication. Combined with the specific register physical addresses pre-allocated to the exhaust fan inverter, fresh air unit inverter, and air conditioning drive board in the system hardware topology, the system performs structured encoding and data segment concatenation and encapsulation to generate a set of speed control decision parameters that can be directly read by the underlying drive unit.

[0041] By employing a rigorous mathematical iterative algorithm, the underlying physical parameters and the mechanical resistance boundary of the pipeline network were self-consistently coupled, avoiding program crashes under extreme conditions and accurately quantifying the optimal drive frequency parameters of each actuator.

[0042] The control signal generation module, based on the speed regulation decision parameter set, generates exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands respectively through preset mapping rules, and integrates and encapsulates them into a control command cluster. Furthermore, in the control signal generation module, the fusion and encapsulation into a control instruction suite specifically includes the following steps: Extract the numerical components of each control node from the speed regulation decision parameter set, classify them according to the preset hardware device address table, and generate a classification parameter matrix. The classification parameter matrix is ​​input into the preset mapping rules to perform control protocol conversion, generating exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands respectively. Extract the exhaust frequency conversion command, fresh air linkage command, and air conditioning adjustment command, and add a system global clock stamp to the exhaust frequency conversion command, fresh air linkage command, and air conditioning adjustment command, and perform timing alignment operation to generate a timing alignment command set; According to the preset underlying bus communication data frame format, the timing alignment instruction set is processed by message packing and checksum filling operations to generate a control instruction set.

[0043] Specifically, after the aforementioned frequency conversion decision evaluation module generates a speed regulation decision parameter set and writes it into the main control chip's memory, the control signal generation module extracts this speed regulation decision parameter set through the internal communication bus. The system performs byte parsing on this parameter set to extract the decimal floating-point target frequency values ​​for each control node. Since the underlying hardware registers only accept integer data, the system first uses a preset precision scaling factor to quantize and convert the above floating-point target frequency values, generating an integer classification parameter matrix. The specific classification matrix construction and quantization parsing algorithm are as follows: ; In the formula, and as well as These represent the target frequency values ​​of the exhaust fan, fresh air unit, and air conditioning compressor extracted from the speed control decision parameter set, respectively. This indicates the preset frequency precision scaling factor; and These represent the physical address of the inverter node classified to the first row of exhaust fan drive dimension and the quantized and converted exhaust fan shaping frequency value, respectively. and These represent the physical address of the inverter node classified to the second row of the fresh air unit drive dimension and the quantized fresh air shaping frequency value, respectively. and These represent the physical address of the communication board node classified to the third row of the air conditioner driver dimension and the quantized converted air conditioner frequency value, respectively. This represents the reconstructed classification parameter matrix. The system then inputs this matrix into a preset mapping rule to perform underlying control protocol conversion. Based on the standard industrial control bus protocol, the system matches the corresponding write register function codes to the control data for each dimension, and generates exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands respectively through an algebraic shift concatenation algorithm. The specific core command mapping concatenation and parsing algorithm is as follows: ; In the formula, This indicates the control dimension index parameter, when When it equals 1, it corresponds to the exhaust fan control dimension. When it equals 2, it corresponds to the control dimension of the fresh air system. When the value is 3, it corresponds to the central air conditioning control dimension; Represents the first element in the classification parameter matrix. The physical address of the node in each dimension; This represents the underlying write register function code constant matched in the preset mapping rules; Represents the corresponding number in the classification parameter matrix. Integer frequency values ​​in each dimension; This represents the first generation generated by shifting and concatenating according to mapping rules. The system generates core control commands for each dimension. Based on these commands, it outputs exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands. To ensure the synchronization of the distributed actuators and eliminate wind pressure surge during fan startup, the system extracts the three generated control commands and retrieves the system global clock stamp of the current main control board. The system appends a unified future trigger timestamp to each of the exhaust frequency conversion command, fresh air linkage command, and air conditioning adjustment command, and performs timing alignment operations to generate a timing alignment command set. The specific timing alignment appending parsing algorithm is as follows: ; In the formula, This indicates the generated first... Core control instructions for each dimension; This represents the current absolute time in milliseconds extracted from the system's global clock stamp; This represents the preset communication transmission bus delay compensation constant. This represents the timing alignment instructions for the corresponding execution dimension, generated by appending timestamps. The timing alignment instructions for exhaust, fresh air, and air conditioning dimensions together form the timing alignment instruction set. Finally, the system performs message packaging operations on each instruction in the timing alignment instruction set according to the preset underlying bus communication data frame format, and generates a cyclic redundancy check code using modulo-2 polynomial division to fill the check bits. The specific message encapsulation and parsing algorithm is as follows: ; In the formula, This refers to the timing alignment instructions generated above; This represents the cyclic redundancy check generator polynomial constant specified by the underlying bus communication protocol. This represents the specific value of the sixteen-bit cyclic redundancy check bit calculated by performing the modulo-2 removal method with the timing alignment instruction as the dividend and the polynomial constant as the divisor. This indicates a standard physical communication data frame generated after fully filling the checksum at the end. The system combines and packages the exhaust data frame, fresh air data frame, and air conditioning data frame into a control command suite and sends it to the transmit buffer of the underlying serial communication port.

[0044] By converting the calculated frequency parameters into compliant electrical communication messages and forcibly aligning the hardware execution times of multiple devices, the accuracy and synchronization of the underlying driver response are ensured.

[0045] The execution module drives the associated exhaust fans, fresh air units, and central air conditioning systems to perform corresponding frequency conversion and cooling capacity adjustment operations based on the control instruction set.

[0046] Furthermore, in the execution module, driving the associated exhaust fan, fresh air unit, and central air conditioning system to perform corresponding frequency conversion and cooling capacity adjustment operations specifically includes the following steps: The control instruction set is disassembled by the underlying protocol parser to extract exhaust frequency conversion instructions, fresh air linkage instructions, and air conditioning adjustment instructions. Establish a logical mapping relationship between exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands and preset physical communication ports, and generate a sequence of commands to be distributed; Based on the encoding format of the exhaust frequency conversion command, fresh air linkage command, and air conditioning adjustment command in the command sequence to be distributed, the exhaust frequency conversion command, fresh air linkage command, and air conditioning adjustment command are converted into underlying physical drive signals that meet the requirements of the hardware communication protocol. The underlying physical drive signals are output to the first frequency converter associated with the exhaust fan, the second frequency converter associated with the fresh air fan, and the cooling capacity regulating valve controller associated with the central air conditioning system to complete the frequency conversion and cooling capacity regulation operations.

[0047] Specifically, after the aforementioned control signal generation module sends the control command suite to the transmission buffer, the execution module takes over the data transmission process through the underlying hardware communication driver subroutine. The system first uses the underlying protocol parser to decompose the various standard physical communication data frames in the control command suite.

[0048] The parser uses a reverse shift masking algorithm to remove the 16-bit cyclic redundancy check code at the end of each data frame, and further performs algebraic separation between the high 32 bits of control instructions and the low 32 bits of timestamps. This allows for the accurate extraction of the core instruction content and corresponding trigger times of exhaust frequency conversion instructions, fresh air linkage instructions, and air conditioning adjustment instructions. The specific frame decomposition and instruction separation parsing algorithm is as follows: ; ; ; In the formula, Indicates the first [number] read from the control instruction set. A standard physical communication data frame in a dimension, when When the values ​​are one, two, and three, they correspond to the dimensions of exhaust ventilation, fresh air ventilation, and air conditioning control, respectively. This indicates the timing alignment instruction obtained after stripping the 16-bit parity bit and restoring the original timing. This represents the number of bits extracted after algebraic shifting thirty-two bits to the right. The core control commands for each dimension are the corresponding exhaust frequency conversion commands, fresh air linkage commands, or air conditioning adjustment commands. This represents the specific value of the preset future trigger timestamp extracted through modulo operation. The system then reads the built-in hardware topology configuration file to establish a logical mapping relationship between the extracted core control instructions and the preset physical communication ports. The system extracts the target hardware address carried in the core control instructions through a high-order right shift operation and compares this address with the asynchronous transceiver interface on the system motherboard to generate a sequence of instructions to be distributed. The specific port addressing mapping parsing algorithm is as follows: ; ; In the formula, This represents the eight-bit physical address of the target hardware node extracted from the core control instructions; A constant representing the base physical starting address of the system's underlying serial port register; A constant representing a fixed memory address offset between registers of different physical communication ports; This represents the address matching return coefficient. When the extracted target hardware node physical address points to the exhaust inverter, the coefficient is one; when it points to the fresh air inverter, the coefficient is two; and when it points to the air conditioning valve controller, the coefficient is three. This represents the result of the mapping calculation used to output the first... The system specifies the physical transmission register address for each dimension instruction. The system writes the core control instructions into the corresponding physical transmission register address, completing the queuing and assembly of the multi-channel instruction sequence to be distributed. After obtaining the instruction sequence to be distributed, the system's main control chip's hardware timer continuously compares the current system global clock with the extracted preset future trigger timestamp.

[0049] At the instant the current clock value is greater than or equal to the trigger timestamp, the system synchronously triggers the transmit interrupt service routine of the underlying communication chip. Based on the digital encoding format of the exhaust frequency conversion command, fresh air linkage command, and air conditioning adjustment command in the sequence, the transceiver hardware converts the above digital sequence into differential level drive signals that meet the requirements of the industrial bus hardware communication protocol, i.e., the underlying physical drive signals. The specific physical digital to differential level conversion algorithm is as follows: ; In the formula, Indicates the first Within the nth transmission bit clock cycle, the sequence of instructions to be distributed is the th... The current binary bit value of the instruction being shifted out; this bit value is either one or zero. This represents a preset positive differential logic voltage constant that meets the relevant industrial bus physical layer standards. This represents a preset negative differential logic voltage constant that meets the relevant industrial bus physical layer standards. This indicates the underlying physical drive differential voltage signal generated during the current bit cycle and output to the twisted-pair bus. The system then synchronously outputs the generated underlying physical drive signal via the physical bus cable to the first inverter associated with the exhaust fan, the second inverter associated with the fresh air unit, and the cooling capacity regulating valve controller associated with the central air conditioning system. Upon receiving the differential voltage signal, the first and second inverters restore it to the target frequency and drive the motor to accelerate or decelerate. The cooling capacity regulating valve controller, upon receiving the signal, drives the valve actuator to change the chilled water flow cross-sectional area, thus comprehensively completing the multi-device coordinated frequency conversion and cooling capacity regulation operation.

[0050] By accurately parsing and converting digital instructions at the software level into electrical drive signals at the hardware level, strict synchronous response of all physical devices under a unified time boundary is achieved, thus closing the overall control process.

[0051] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fan energy-saving intelligent control system based on equipment detection, characterized in that, include: The working condition sensing and acquisition module is used to acquire the oil fume heat signal in the cooking area, the continuous micro pressure difference signal on both sides of the exhaust filter, the operating current signal of the exhaust fan, and the temperature and humidity sensing signals of the indoor and outdoor air outlets, and generate the original physical and environmental dataset. The resistance state extraction constraint module, based on the original physical and environmental dataset, analyzes the spatiotemporal coupling relationship between the continuous micro-pressure difference signal and the operating current signal of the exhaust fan through a fluid dynamics model, extracts the pipeline resistance index, and aligns and compares the pipeline resistance index with a preset baseline to generate a pipeline frequency limiting constraint matrix. The environmental enthalpy calculation and analysis module, based on the temperature and humidity sensing signals of the indoor and outdoor air outlets in the original physical and environmental dataset, uses the enthalpy-humidity map algorithm model to perform thermodynamic calculations and outputs indoor and outdoor environmental enthalpy vectors. The feature fusion coupling module receives the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector, calculates the correlation factor between the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector through feature splicing and weight verification, and generates a global ratio descriptor. The frequency conversion decision evaluation module calculates the target exhaust baseline volume based on the global ratio descriptor and the oil fume thermal signal in the original physical and environmental dataset, and generates a speed regulation decision parameter set through multivariate nonlinear solution. The control signal generation module generates exhaust frequency conversion commands, fresh air linkage commands, and air conditioning adjustment commands respectively based on the speed regulation decision parameter set and through preset mapping rules, and integrates and encapsulates them into a control command cluster. The execution module drives the associated exhaust fan, fresh air fan, and central air conditioning system to perform corresponding frequency conversion and cooling capacity adjustment operations based on the control command set.

2. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the aforementioned operational condition sensing and acquisition module, the generation of the original physical and environmental dataset specifically includes the following steps: The cooking area's oil fume heat signal, the continuous micro-pressure difference signal on both sides of the exhaust filter, the exhaust fan's operating current signal, and the indoor and outdoor air vents' temperature and humidity sensing signals are collected by a preset sampling frequency, and then processed by analog-to-digital conversion to generate a discrete digital signal sequence. The arrival timestamps of the discrete digital signal sequence are extracted, the relative time offset is calculated based on the system global clock, and the discrete digital signal sequence is time-aligned by an interpolation resampling algorithm to generate a synchronization multidimensional state matrix. The synchronous multidimensional state matrix is ​​input into a preset filtering model, and the high-frequency environmental noise components in the synchronous multidimensional state matrix are separated and removed to output the filtering feature matrix. Extract each data dimension vector from the filtered feature matrix, and according to a preset data encapsulation protocol, structurally concatenate each data dimension vector along with the corresponding timestamp index and sensor node identifier to generate the original physical and environmental dataset.

3. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the resistance state extraction constraint module, the generation of the pipeline network frequency limiting constraint matrix specifically includes the following steps: The continuous micro-pressure difference signal and the operating current signal of the exhaust fan are extracted from the original physical and environmental dataset, and the features are spliced ​​according to the time series dimension to generate resistance feature vector pairs. The resistance feature vector is input into a preset fluid dynamics model, and the spatiotemporal coupling coefficient between the continuous micro-pressure difference signal and the operating current signal of the exhaust fan is calculated through polynomial fitting, and the pipeline resistance index is output. Extract the pipeline resistance index, align the pipeline resistance index with a preset baseline and perform differential operation to generate a resistance attenuation deviation that characterizes the current pipeline blockage state; The resistance attenuation deviation is input into a preset frequency limiting mapping function to calculate the upper limit threshold of the frequency constraint. The upper limit threshold of the frequency constraint is then reconstructed into a matrix according to the system control dimension to generate the pipeline frequency limiting constraint matrix.

4. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the environmental enthalpy calculation and analysis module, the output of indoor and outdoor environmental enthalpy vectors specifically includes the following steps: The temperature and humidity sensing signals of the indoor and outdoor air outlets are extracted from the original physical and environmental dataset and classified and packaged according to the spatial identifiers of indoor and outdoor spaces to generate a set of environmental thermal parameters. The environmental thermal parameter component set is input into the enthalpy-humidity diagram algorithm model. The specific enthalpy values ​​of indoor and outdoor measuring points at the current sampling time are obtained by solving the preset air state equation, and the instantaneous specific enthalpy data pairs are output. Perform time-domain filtering based on a sliding window on the instantaneous enthalpy data pairs, calculate the statistical mean of enthalpy values ​​within a preset sampling period, and generate a mean enthalpy value sequence; According to the preset vector construction rules, the mean enthalpy sequence is mapped to the corresponding multidimensional spatial coordinate system to generate indoor and outdoor environmental enthalpy vectors.

5. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the feature fusion coupling module, the generation of the global ratio descriptor specifically includes the following steps: Obtain the network frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vectors, perform data dimension alignment and feature splicing operations, and generate a cross-domain feature fusion matrix; The cross-domain feature fusion matrix is ​​input into a preset weight allocation model to perform weight verification operations and output a weighted feature sequence. Perform covariance analysis on the weighted feature sequence to calculate and extract the numerical correlation factor between the pipeline frequency limiting constraint matrix and the indoor and outdoor environmental enthalpy vector; According to the preset format encapsulation protocol, the numerical correlation factors are subjected to multidimensional coordinate mapping and structured recombination to generate a global ratio descriptor.

6. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the variable frequency decision evaluation module, the generation of the speed regulation decision parameter set specifically includes the following steps: The oil fume thermal signal is extracted from the original physical and environmental dataset, and the oil fume thermal signal is input into a preset load mapping function for numerical conversion to calculate the target exhaust baseline volume. Obtain the global ratio descriptor, align the target exhaust baseline quantity with the global ratio descriptor using data dimension alignment and multi-dimensional feature fusion, and generate the working condition state vector to be solved. The state vector of the working condition to be solved is input into a preset multivariable nonlinear equation system to perform iterative analytical calculations and output the target frequency values ​​of each control node. According to the preset hardware control protocol, the target frequency value is structured and encapsulated to generate a speed regulation decision parameter set.

7. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the control signal generation module, the fusion and encapsulation into a control instruction cluster specifically includes the following steps: Extract the numerical components of each control node in the speed regulation decision parameter set, and classify them according to the preset hardware device address table to generate a classification parameter matrix. The classification parameter matrix is ​​input into the preset mapping rule to perform control protocol conversion, generating exhaust frequency conversion command, fresh air linkage command and air conditioning adjustment command respectively; Extract the exhaust frequency conversion command, the fresh air linkage command, and the air conditioning adjustment command, and add a system global clock stamp to the exhaust frequency conversion command, the fresh air linkage command, and the air conditioning adjustment command, and perform timing alignment operation to generate a timing alignment command set; According to the preset underlying bus communication data frame format, the timing alignment instruction set is processed by message packing and checksum filling operations to generate a control instruction set.

8. The wind turbine energy-saving intelligent control system based on equipment detection according to claim 1, characterized in that, In the execution module, the drive-associated exhaust fan, fresh air unit, and central air conditioning system perform corresponding frequency conversion and cooling capacity adjustment operations, specifically including the following steps: The control instruction set is disassembled by the underlying protocol parser to extract the exhaust frequency conversion instruction, the fresh air linkage instruction, and the air conditioning adjustment instruction; Establish a logical mapping relationship between the exhaust frequency conversion command, the fresh air linkage command, and the air conditioning adjustment command and the preset physical communication port, and generate a sequence of commands to be distributed; According to the encoding format of the exhaust frequency conversion instruction, the fresh air linkage instruction, and the air conditioning adjustment instruction in the instruction sequence to be distributed, the exhaust frequency conversion instruction, the fresh air linkage instruction, and the air conditioning adjustment instruction are converted into underlying physical drive signals that meet the hardware communication protocol requirements. The underlying physical drive signals are output to the first frequency converter associated with the exhaust fan, the second frequency converter associated with the fresh air fan, and the cooling capacity regulating valve controller associated with the central air conditioning system, respectively, to complete the frequency conversion and cooling capacity regulation operations.

Images