A method and system for intelligent optimization of distributed building heating parameters

Abstract

This invention discloses a method and system for intelligent optimization of heating parameters in distributed buildings. The method uses an improved SSD algorithm to extract multi-scale features from the heating area, establishes a dynamic population distribution data sequence using a heating time-series and population optimization model, collects and normalizes heating system parameters, inputs the data into an optimization decision model, and outputs an optimized combination scheme through a multi-layer neural network. Heating parameters are adjusted in real time based on this, and the operating data is fed back to the optimization model at a set period. The system includes data acquisition, feature extraction, time-series and population analysis, optimization decision-making, parameter adjustment, and feedback update units, all of which work collaboratively. This method and system can fully consider dynamic changes in the environment and population, deeply explore parameter correlations, achieve intelligent optimization of heating parameters, and improve heating performance and energy efficiency.

Description

Technical Field

[0001] This invention relates to the field of building heating, and in particular to a method and system for intelligent optimization of decentralized building heating parameters. Background Technology

[0002] With the development of the construction industry and the increasing demand for energy, decentralized building heating systems are widely used in many building scenarios due to their flexibility and applicability. However, traditional decentralized building heating systems have many problems in parameter optimization, making it difficult to meet the current needs for high efficiency, energy saving, and precise control.

[0003] In existing technologies, on the one hand, traditional methods for adjusting heating parameters often rely on manual experience or simple preset rules, lacking a comprehensive consideration of dynamic factors such as the building environment and population flow. For example, they cannot adjust the operating parameters of heating equipment in a timely and accurate manner based on changes in the number of people in the building at different times and fluctuations in indoor and outdoor temperatures, resulting in poor heating performance and significant energy waste. On the other hand, the algorithm models used in traditional methods are relatively simple and outdated, failing to fully explore the inherent relationships and potential patterns between various parameters of the heating system. When dealing with complex and ever-changing heating scenarios, they struggle to achieve intelligent optimization of heating parameters and cannot achieve an ideal balance between heating performance and energy efficiency. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a method and system for intelligent optimization of decentralized building heating parameters.

[0005] The technical solution adopted in this invention is an intelligent optimization method for decentralized building heating parameters, comprising the following steps:

[0006] Step S1: Based on the improved SSD algorithm, perform multi-scale feature extraction on the heating area of ​​the distributed building to obtain the feature data set of the heating area under different spatial dimensions;

[0007] Step S2: Using the heating time series and population optimization model, combined with historical heating data and historical population flow data within buildings, establish the correlation mapping relationship between heating time series and population distribution to obtain a dynamic data sequence of population distribution changes over time.

[0008] Step S3: Collect various parameters of the decentralized building heating system, including the power parameters of the heating equipment, the thermal conductivity parameters of the pipes, and the indoor and outdoor temperature parameters. Normalize the collected parameters to form a standardized parameter vector.

[0009] Step S4: Input the feature data set, dynamic data sequence and standardized parameter vector into the optimization decision model, and output the optimized combination scheme of heating parameters through the multi-layer neural network calculation of the optimization decision model;

[0010] Step S5: Based on the optimized combination scheme, adjust the parameters of the decentralized building heating system in real time and record the adjusted heating system operation data simultaneously;

[0011] Step S6: Feed the running data back to the optimization decision model at set time intervals, iteratively update the optimization decision model, and continuously optimize the combination scheme of heating parameters.

[0012] Furthermore, the optimization decision model includes heating parameter optimization formulas:

[0013]

[0014] Among them, P opt The optimized heating parameter combination scheme; α is the weighting adjustment coefficient; n is the number of data samples; X i f represents the feature data extracted from the i-th heating zone using the improved SSD algorithm; SSD (X i To improve the SSD algorithm for feature data X i The processing function for Y; i This represents dynamic data on the population distribution within a building during the i-th time period; g HTPOM (Y i The heating time series and population optimization model uses population dynamic data Y. i The processing function for Z; i h represents the standardized heating parameter vector at the i-th sampling time. para (Z i For the standardized parameter vector Z i Parameter processing functions.

[0015] Furthermore, in step S4, the optimization decision model also includes a dynamic adjustment formula for heating parameters:

[0016]

[0017] Where ΔP is the adjustment amount of the heating parameters; β is the time sensitivity coefficient; To optimize scheme P opt Rate of change over time; γ pop The population mobility impact coefficient reflects the degree to which changes in population distribution affect heating parameters; δ para This is the comprehensive influence coefficient of heating parameters, reflecting the coupling effect between various heating parameters.

[0018] Furthermore, in step S1, when the improved SSD algorithm extracts features from the heating area, it employs a multi-scale sliding window mechanism to calculate the feature response values ​​at different scales using the following formula:

[0019]

[0020] Among them, R scale The feature response value is at a preset scale; m is the number of feature dimensions; w j The weight of the j-th feature dimension; To analyze the characteristic data X of the heating area at a preset scale region The transformation function.

[0021] Further, in step S2, the heating timing and population optimization model constructs a population distribution prediction formula:

[0022]

[0023] in, θ represents the predicted population distribution over the next k time periods; s is the length of the historical data selection; θ l Historical population data Y t-l The weighting coefficients; ∈ pop This represents the error term in the population distribution prediction.

[0024] Furthermore, in step S3, when normalizing the collected heating parameters, the following formula is used:

[0025]

[0026] Among them, Z norm Z represents the normalized parameter values; Z represents the original parameter values; Z max Z represents the maximum value of this parameter in historical data. min This is the minimum value of the parameter in historical data.

[0027] Furthermore, step S3 includes the following sub-steps:

[0028] Step S3.1: Use a high-precision sensor array to collect the power parameters of the heating equipment in real time. Through the multi-channel synchronous sampling technology of the sensor, ensure the timeliness and completeness of the power parameter collection.

[0029] Step S3.2: Using a thermal conductivity coefficient detection device, the thermal conductivity coefficient parameter of the pipeline is measured based on the principle of heat flow meter to obtain the thermal conductivity performance data of the pipeline under different environmental conditions;

[0030] Step S3.3: By using temperature sensors placed inside and outside the building, indoor and outdoor temperature parameters are periodically collected according to the set sampling frequency to form a continuous temperature data sequence;

[0031] Step S3.4: The collected power parameters of heating equipment, thermal conductivity parameters of pipes, and indoor and outdoor temperature parameters are processed to unify the data format and convert them into a standard data format that meets the system requirements.

[0032] Furthermore, step S4 includes the following sub-steps:

[0033] Step S4.1: Perform data fusion processing on the feature data set, dynamic data sequence, and standardized parameter vector, and construct a complete input data matrix through feature concatenation and temporal alignment techniques;

[0034] Step S4.2: Input the input data matrix into the input layer of the optimization decision model. The data is processed by the convolutional layer in the model to extract features and reduce dimensionality, thereby enhancing the expressive power of the data features.

[0035] Step S4.3: By optimizing the recurrent neural network layer of the decision model, the temporal characteristics of the data are learned and memorized to capture the pattern of heating parameters changing over time;

[0036] Step S4.4: The output layer of the optimization decision model calculates the processed data, outputs an optimized combination scheme of heating parameters, and performs data format conversion to meet the execution requirements of the heating system.

[0037] Further, step S5 includes the following sub-steps:

[0038] Step S5.1: Based on the optimized combination scheme, generate the corresponding parameter adjustment instruction set, and clarify the adjustment target value and adjustment sequence of each heating equipment parameter;

[0039] Step S5.2: Send the parameter adjustment command set to the controller of the heating system through the communication module, and use an encrypted communication protocol to ensure the security and accuracy of command transmission;

[0040] Step S5.3: The heating system controller adjusts the parameters of the heating equipment parameter by parameter according to the received instruction set, and monitors the equipment operating status in real time during the adjustment process;

[0041] Step S5.4: After the parameters are adjusted, start the data recording module to collect and store the adjusted heating system operation data in all aspects, including equipment operation parameters and indoor and outdoor environmental parameters.

[0042] A decentralized building heating parameter intelligent optimization system includes:

[0043] The data acquisition unit is used to collect various parameters of the decentralized building heating system, including heating equipment power parameters, pipe thermal conductivity parameters, and indoor and outdoor temperature parameters.

[0044] The feature extraction unit is connected to the data acquisition unit and performs multi-scale feature extraction on the heating area based on the improved SSD algorithm to obtain a feature data set.

[0045] The time series and population analysis unit is connected to the feature extraction unit. Using the heating time series and population optimization model, it establishes the correlation mapping relationship between heating time series and population distribution to obtain dynamic data sequence of population distribution.

[0046] The optimization decision unit is connected to the feature extraction unit, the time series and population analysis unit, and the data acquisition unit, respectively. It performs calculations based on the input data and outputs an optimized combination scheme of heating parameters.

[0047] The parameter adjustment unit is connected to the optimization decision unit and adjusts the heating system parameters in real time according to the optimization combination scheme.

[0048] The feedback update unit is connected to the parameter adjustment unit and the optimization decision unit, and feeds back the adjusted heating system operation data to the optimization decision unit to iteratively update the optimization decision model.

[0049] Beneficial Effects: This invention proposes a method and system for intelligent optimization of distributed building heating parameters. It utilizes an improved SSD algorithm to extract multi-scale features from the heating area and combines heating time series and population optimization models to correlate historical population flow data within the building with the heating time series. This accurately captures dynamic changes in indoor and outdoor environments and population distribution, enabling heating parameter adjustments to move beyond fixed rules and optimize in real-time and intelligently based on actual scenarios. To address the shortcomings of traditional single-model and lagging algorithms, this system constructs multiple innovative model formulas. The optimization decision model uses multi-layer neural network calculations, combined with heating parameter optimization formulas and dynamic adjustment formulas, to deeply explore the potential relationships between various parameters of the heating system, achieving efficient processing of complex heating scenarios. Meanwhile, the system forms a closed-loop mechanism of "data acquisition - feature extraction - analysis and decision-making - parameter adjustment - feedback and update". The collected parameters are normalized and then input into the optimization decision-making model. The output optimized combination scheme is used to adjust the heating system parameters in real time. The adjusted operating data is then fed back to the model for iterative updates, continuously improving the heating effect and energy utilization efficiency. This ensures indoor heating comfort and significantly reduces energy waste, bringing an efficient and intelligent solution to the field of decentralized building heating. Attached Figure Description

[0050] Figure 1 This is a flowchart of the method steps of the present invention;

[0051] Figure 2 This is a diagram showing the system unit composition of the present invention. Detailed Implementation

[0052] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] like Figure 1 As shown, a method for intelligent optimization of heating parameters in decentralized buildings includes the following steps:

[0054] Step S1: Based on the improved SSD algorithm, perform multi-scale feature extraction on the heating area of ​​the distributed building to obtain the feature data set of the heating area under different spatial dimensions;

[0055] Specifically, this step employs an improved SSD (Single Shot Detector) algorithm to perform multi-scale feature extraction on the heating areas of distributed buildings. During implementation, the algorithm scans and analyzes the heating area from different spatial dimensions according to pre-defined scale specifications. For example, from a tiny room corner to an entire floor space, or even the entire heating area of ​​a building, the algorithm can traverse windows of different scales, thereby capturing environmental, structural, and other feature information at each spatial level. This feature information covers heating-related spatial parameters such as room layout, wall insulation performance, and the orientation and airtightness of doors and windows. Through this comprehensive and detailed multi-scale scanning, a feature data set of the heating area containing rich spatial dimensional characteristics is finally obtained, providing important basic data support for subsequent heating parameter optimization.

[0056] Multi-scale feature extraction can fully consider the complex and varied spatial characteristics of decentralized building heating areas. Feature data extracted at different scales can reflect heating influencing factors at different levels. Large-scale features help to grasp the overall heating layout and environment, while small-scale features can accurately locate the specific needs and potential problems of local areas. This comprehensive feature dataset enables subsequent optimization decisions to be based on more accurate and complete information, avoiding optimization biases caused by considering only a single-scale feature, thus laying a solid foundation for achieving more efficient and accurate heating parameter optimization.

[0057] Step S2: Using the heating time series and population optimization model, combined with historical heating data and historical population flow data within buildings, establish the correlation mapping relationship between heating time series and population distribution to obtain a dynamic data sequence of population distribution changes over time.

[0058] Specifically, this step involves in-depth mining and analysis of historical heating data and historical population flow data within buildings through a heating time series and population optimization model. During implementation, the model subdivides the time dimension, extracting and integrating heating and population flow data at each specific time point (e.g., every hour, every half hour). Historical heating data includes information such as the operating parameters of heating equipment and changes in indoor and outdoor temperatures over different time periods; historical population flow data records the entry and exit times, dwell time, and number of people gathered in various areas of the building. Through the analysis and processing of this massive amount of historical data, the model can discover the inherent relationship between heating demand and population distribution, thereby establishing a correlation mapping between the two. Based on this mapping relationship, the model can predict the dynamic changes in population distribution in various areas of the building at different future time points, ultimately generating a dynamic data sequence of population distribution changes over time.

[0059] Population distribution is a key factor influencing heating demand. The activity patterns and areas of concentration of people within buildings change over time, leading to corresponding changes in heating demand. By establishing a mapping relationship between heating time series and population distribution, the impact of dynamic population changes on heating demand can be accurately grasped. This prevents the heating system from operating blindly; instead, it allows for advance planning and adjustment of heating parameters based on dynamic population distribution data. Sufficient heat supply can be provided in densely populated areas and during specific times, while heating intensity can be appropriately reduced in sparsely populated areas and during less populated times. This achieves on-demand heating, improving the targeting and effectiveness of heating, while also helping to reduce energy consumption and improve energy efficiency.

[0060] Step S3: Collect various parameters of the decentralized building heating system, including the power parameters of the heating equipment, the thermal conductivity parameters of the pipes, and the indoor and outdoor temperature parameters. Normalize the collected parameters to form a standardized parameter vector.

[0061] Specifically, in this step, various specialized sensor devices are first used to collect key parameters of the decentralized building heating system in real time. For heating equipment power parameters, high-precision power sensors are installed at key circuit nodes of the heating equipment to monitor power consumption during operation. For pipe thermal conductivity parameters, a detection device based on the principle of a heat flow meter is used, placed at different locations on the pipes to measure the thermal conductivity of the pipes under different ambient temperatures and fluid conditions. Indoor and outdoor temperature parameters are collected periodically by temperature sensors distributed throughout the building, at a set sampling frequency (e.g., once per minute). After collecting these parameters, due to the different numerical ranges and dimensions of each parameter, normalization is required to facilitate subsequent data processing and analysis. Normalization maps all parameter values ​​to a unified numerical range (usually [0, 1] or [-1, 1]), eliminating differences in dimensions and numerical magnitude, ultimately forming a standardized parameter vector, making different types of parameters comparable and fusionable.

[0062] The operational status of decentralized building heating systems is determined by numerous different types of parameters. Only by comprehensively and accurately collecting these parameters and processing them appropriately can the actual operating conditions of the system be accurately reflected. Normalization processing allows all types of parameters to be analyzed and calculated under the same standard, avoiding calculation biases and decision-making errors caused by excessive parameter differences. Standardized parameter vectors serve as crucial input data for subsequent optimization decision models; their accuracy and standardization directly affect the quality and reliability of the optimization scheme, providing effective data support for achieving precise heating parameter optimization.

[0063] Step S4: Input the feature data set, dynamic data sequence and standardized parameter vector into the optimization decision model, and output the optimized combination scheme of heating parameters through the multi-layer neural network calculation of the optimization decision model;

[0064] Specifically, this step uses the feature data set, population distribution dynamic data sequence, and standardized parameter vector obtained in the previous three steps as input data for the optimization decision model. The optimization decision model is built on a multi-layer neural network architecture, which includes an input layer, multiple hidden layers, and an output layer. When data is input into the model's input layer, it is processed sequentially through each hidden layer. In the hidden layers, neurons use specific activation functions to calculate and transform the input data, continuously extracting key features and inherent patterns from the data, achieving feature extraction, dimensionality reduction, and information fusion. As data is passed and processed layer by layer in the multi-layer neural network, the model gradually learns and understands the spatial characteristics of the heating area, dynamic population changes, and the complex relationships between heating system parameters. Finally, after calculation at the output layer, the model outputs an optimized combination of heating parameters based on the learned knowledge and patterns, specifying the optimized values ​​for specific parameters such as the operating power of heating equipment, pipeline flow control, and temperature settings.

[0065] The optimized decision-making model, through comprehensive analysis and in-depth processing of multi-source data, overcomes the limitations and one-sidedness of traditional methods for adjusting heating parameters. The powerful learning and computational capabilities of multi-layer neural networks enable them to uncover complex hidden relationships between data points, fully considering the impact of spatial environment, population changes, and system parameters on heating performance. The output optimized combination scheme is based on comprehensive and accurate data analysis. Compared to traditional experience-based or simple rule-based parameter adjustment methods, it can more scientifically and rationally optimize heating parameters, effectively improving the operating efficiency and heating quality of the heating system, achieving the dual goals of energy saving and comfort.

[0066] Step S5: Based on the optimized combination scheme, adjust the parameters of the decentralized building heating system in real time and record the adjusted heating system operation data simultaneously;

[0067] Specifically, after obtaining the optimized combination of heating parameters output by the optimized decision-making model, this step immediately executes real-time adjustments to the parameters of the decentralized building heating system. In practice, the optimized combination scheme is converted into corresponding control commands via a communication module and transmitted to each controller in the heating system using an encrypted communication protocol. These controllers are responsible for controlling different heating equipment and components, such as heating boilers, circulating water pumps, and valves. Upon receiving the commands, the controllers adjust the operating parameters of the heating equipment one by one according to the instructions, such as adjusting the combustion power of the heating boiler, changing the speed of the circulating water pump to control the pipeline flow, and adjusting the valve opening to regulate heat distribution. Simultaneously with parameter adjustments, the system activates the data recording module, using various sensors and data acquisition devices to collect comprehensive, real-time data on the adjusted heating system operation. The collected data covers the actual operating parameters of the heating equipment (such as power, temperature, and flow rate), indoor and outdoor environmental parameters (such as temperature, humidity, and wind speed), and system energy consumption data. This data is stored to form a complete operational data record.

[0068] Transforming the theoretical optimization schemes output by the optimized decision-making model into actual heating system operation control represents a significant shift from data analysis to practical operation. Real-time adjustments ensure that the heating system responds promptly to changes in the environment and demand, quickly reaching the optimized operating state and improving heating efficiency. Synchronously recording the adjusted operating data provides a reliable basis for subsequent system optimization and evaluation. Analyzing this operating data allows us to understand the actual effectiveness of the optimization scheme, identify potential problems and shortcomings, and provide data support for further improvement and optimization of heating parameters, forming a closed-loop management model for continuous optimization and ensuring the heating system always operates efficiently and stably.

[0069] Step S6: Feed the running data back to the optimization decision model at set time intervals, iteratively update the optimization decision model, and continuously optimize the combination scheme of heating parameters.

[0070] Specifically, this step feeds back the heating system operation data recorded in step S5 to the optimization decision model according to a pre-set fixed time period (such as hourly, daily, etc.). The operation data contains various real-world information about the heating system during actual operation, reflecting the effectiveness and existing problems of the current optimization scheme in practical application. After the operation data is input into the optimization decision model, the model adjusts and optimizes its parameters and structure based on this new data, i.e., it iterates and updates. During the iteration process, the model relearns and analyzes the characteristics and patterns in the data, compares the previous learning results, discovers new changes and trends in the data, and adjusts the neuron connection weights, activation function parameters, etc., within the model to better adapt to the constantly changing heating environment and demands. After iterative updates, the optimization decision model, when receiving new input data, can output more accurate and optimized combinations of heating parameters, achieving continuous optimization of heating parameters and continuously improving the operating performance and energy efficiency of the heating system.

[0071] The heating environment in decentralized buildings is dynamic, influenced by changes in occupant activity patterns, fluctuations in weather conditions, and degradation of equipment performance. By periodically feeding operational data back to the optimization decision model for iterative updates, the model can remain adaptable and sensitive to current realities. This prevents optimization schemes from failing due to model lag, ensuring that the heating system can adjust heating parameters promptly according to actual needs under various complex and changing environments, achieving efficient, energy-saving, and comfortable heating goals. This continuous optimization mechanism provides strong support for the long-term stable operation and performance improvement of decentralized building heating systems.

[0072] Preferably, the optimization decision model includes heating parameter optimization formulas:

[0073]

[0074] Among them, P opt The optimized heating parameter combination scheme; α is the weighting adjustment coefficient; n is the number of data samples; X i f represents the feature data extracted from the i-th heating zone using the improved SSD algorithm; SSD (X i To improve the SSD algorithm for feature data X i The processing function for Y; i This represents dynamic data on the population distribution within a building during the i-th time period; g HTPOM (Y i The heating time series and population optimization model uses population dynamic data Y. i The processing function for Z; i h represents the standardized heating parameter vector at the i-th sampling time. para (Zi For the standardized parameter vector Z i Parameter processing functions.

[0075] Specifically, the optimization formula for heating parameters in the decision-making model comprehensively considers multiple factors to generate an optimized combination of heating parameters. In terms of implementation, a weighted summation operation is performed on the heating area feature data processed by the improved SSD algorithm, the population dynamic data output from the heating time series and population optimization model, and the standardized heating parameter vector. The weight adjustment coefficient is set according to actual needs to balance the influence of different data; the number of data samples determines the amount of data involved in the calculation. Its significance lies in closely integrating spatial characteristics, population changes, and system parameters, and deriving a scientifically reasonable combination of heating parameters through quantitative calculation. This ensures that the operating parameters of the heating system accurately match actual needs, avoiding energy waste and poor heating performance.

[0076] Preferably, in step S4, the optimization decision model further includes a dynamic adjustment formula for heating parameters:

[0077]

[0078] Where ΔP is the adjustment amount of the heating parameters; β is the time sensitivity coefficient; To optimize scheme P opt Rate of change over time; γ pop The population mobility impact coefficient reflects the degree to which changes in population distribution affect heating parameters; δ para This is the comprehensive influence coefficient of heating parameters, reflecting the coupling effect between various heating parameters.

[0079] Specifically, the dynamic adjustment formula for heating parameters determines the real-time adjustment amount of these parameters. In implementation, the heating parameters are dynamically adjusted based on the time sensitivity coefficient, the rate of change of the optimized scheme over time, the population flow influence coefficient, and the comprehensive influence coefficient of the heating parameters. The time sensitivity coefficient reflects the system's responsiveness to changes in time; the rate of change of the optimized scheme reflects the dynamic trend of heating demand; the population flow influence coefficient highlights the impact of changes in population distribution on heating; and the comprehensive influence coefficient considers the coupling relationship between various heating parameters. Its significance lies in enabling the heating system to adjust heating parameters in real-time and accurately based on the passage of time, population flow, and interactions between parameters, thereby improving the dynamic adaptability and energy efficiency of the heating system.

[0080] Preferably, in step S1, when the improved SSD algorithm extracts features from the heating area, it employs a multi-scale sliding window mechanism and calculates the feature response values ​​at different scales using the following formula:

[0081]

[0082] Among them, R scale The feature response value is at a preset scale; m is the number of feature dimensions; w j The weight of the j-th feature dimension; To analyze the characteristic data X of the heating area at a preset scale region The transformation function.

[0083] Specifically, the improved SSD algorithm feature extraction process in step S1 employs a multi-scale sliding window mechanism to calculate feature response values. During implementation, at different scales, the feature data of the heating area are transformed and calculated based on the weights of each feature dimension. The number of feature dimensions covers various heating-related factors such as spatial structure and building materials. The weights are set according to the importance of each feature to the heating effect, and the transformation function performs the transformation processing of the feature data. This mechanism can comprehensively capture the feature information of the heating area at different scales, providing rich and accurate basic data for subsequent heating parameter optimization, ensuring that optimization decisions fully consider the impact of spatial factors on heating.

[0084] Preferably, in step S2, the heating timing and population optimization model constructs a population distribution prediction formula:

[0085]

[0086] in, θ represents the predicted population distribution over the next k time periods; s is the length of the historical data selection; θ l Historical population data Y t-l The weighting coefficients; ∈ pop This represents the error term in the population distribution prediction.

[0087] Specifically, the population distribution prediction formula in the heating timing and population optimization model in step S2 predicts future population distribution using historical population data. In implementation, a certain length of historical population data is selected, and a weighted sum is calculated based on the weighting coefficients corresponding to each historical data point. A prediction error term is then added to obtain the predicted future population distribution value. The length of the selected historical data determines the time span of the reference data, the weighting coefficients reflect the degree of influence of data from different historical moments on the prediction results, and the error term is used to correct prediction deviations. This formula can predict dynamic population changes based on historical patterns, allowing the heating system to be planned in advance, adjusting heating parameters according to population distribution trends, achieving on-demand heating, and improving the accuracy of heating and energy efficiency.

[0088] Preferably, in step S3, when normalizing the collected heating parameters, the following formula is used:

[0089]

[0090] Among them, Z norm Z represents the normalized parameter values; Z represents the original parameter values; Z max Z represents the maximum value of this parameter in historical data. min This is the minimum value of the parameter in historical data.

[0091] Specifically, the heating parameter normalization formula in step S3 is used to convert the collected raw heating parameters into standardized data. During implementation, based on the maximum and minimum values ​​of each parameter in historical data, a linear transformation is performed on the raw parameters, mapping them to a specific interval. The normalized parameters eliminate differences in units and numerical magnitudes, facilitating subsequent data processing and analysis. This process ensures the comparability and fusion of heating parameters of different types and magnitudes, providing a guarantee for the optimization decision model to accurately process data and output reliable optimization solutions, thereby improving the overall data processing efficiency and optimization effect of the system.

[0092] Preferably, step S3 includes the following sub-steps:

[0093] Step S3.1: Use a high-precision sensor array to collect the power parameters of the heating equipment in real time. Through the multi-channel synchronous sampling technology of the sensor, ensure the timeliness and completeness of the power parameter collection.

[0094] Step S3.2: Using a thermal conductivity coefficient detection device, the thermal conductivity coefficient parameter of the pipeline is measured based on the principle of heat flow meter to obtain the thermal conductivity performance data of the pipeline under different environmental conditions;

[0095] Step S3.3: By using temperature sensors placed inside and outside the building, indoor and outdoor temperature parameters are periodically collected according to the set sampling frequency to form a continuous temperature data sequence;

[0096] Step S3.4: The collected power parameters of heating equipment, thermal conductivity parameters of pipes, and indoor and outdoor temperature parameters are processed to unify the data format and convert them into a standard data format that meets the system requirements.

[0097] Specifically, firstly, a high-precision sensor array combined with multi-channel synchronous sampling technology is used to collect power parameters of the heating equipment, ensuring the real-time nature and integrity of the data. Secondly, a detection device based on the principle of heat flux is used to measure the thermal conductivity parameters of the pipes, obtaining performance data under different environments. Then, temperature sensors deployed inside and outside the building collect indoor and outdoor temperature parameters at a set frequency, forming a continuous data sequence. Finally, the collected parameters are processed to conform to system requirements. This step-by-step, comprehensive, and accurate collection and processing of heating system parameters provides an accurate and standardized data foundation for subsequent optimization decisions.

[0098] Preferably, step S4 includes the following sub-steps:

[0099] Step S4.1: Perform data fusion processing on the feature data set, dynamic data sequence, and standardized parameter vector, and construct a complete input data matrix through feature concatenation and temporal alignment techniques;

[0100] Step S4.2: Input the input data matrix into the input layer of the optimization decision model. The data is processed by the convolutional layer in the model to extract features and reduce dimensionality, thereby enhancing the expressive power of the data features.

[0101] Step S4.3: By optimizing the recurrent neural network layer of the decision model, the temporal characteristics of the data are learned and memorized to capture the pattern of heating parameters changing over time;

[0102] Step S4.4: The output layer of the optimization decision model calculates the processed data, outputs an optimized combination scheme of heating parameters, and performs data format conversion to meet the execution requirements of the heating system.

[0103] Specifically, the process begins by fusing feature datasets, dynamic data sequences, and standardized parameter vectors, constructing a complete input matrix through feature concatenation and temporal alignment. This matrix is ​​then input into the optimization decision model, where convolutional layers extract features and reduce dimensionality to enhance data representation. Next, recurrent neural network layers learn the temporal characteristics of the data to capture the changing patterns of heating parameters. Finally, the output layer calculates and outputs the optimized combination scheme and converts the data format. These steps work collaboratively, enabling the optimization decision model to fully process multi-source data, deeply mine data value, and output scientifically sound and reasonable heating parameter optimization schemes.

[0104] Preferably, step S5 includes the following sub-steps:

[0105] Step S5.1: Based on the optimized combination scheme, generate the corresponding parameter adjustment instruction set, and clarify the adjustment target value and adjustment sequence of each heating equipment parameter;

[0106] Step S5.2: Send the parameter adjustment command set to the controller of the heating system through the communication module, and use an encrypted communication protocol to ensure the security and accuracy of command transmission;

[0107] Step S5.3: The heating system controller adjusts the parameters of the heating equipment parameter by parameter according to the received instruction set, and monitors the equipment operating status in real time during the adjustment process;

[0108] Step S5.4: After the parameters are adjusted, start the data recording module to collect and store the adjusted heating system operation data in all aspects, including equipment operation parameters and indoor and outdoor environmental parameters.

[0109] Specifically, the process begins by generating a parameter adjustment instruction set based on the optimized combination scheme, clarifying the adjustment targets and sequence for each device's parameters. Then, the instruction set is sent to the heating system controller via a communication module using an encrypted protocol, ensuring secure and accurate transmission. Next, the controller adjusts the heating equipment parameter by parameter according to the instructions and monitors the equipment status in real time. Finally, after the adjustments are completed, the data recording module is activated to collect and store system operation data. This process ensures that the optimization scheme can be accurately and securely implemented in the heating system and provides data support for subsequent system evaluation and optimization, achieving dynamic optimization and efficient operation of the heating system.

[0110] like Figure 2 As shown, a decentralized building heating parameter intelligent optimization system includes:

[0111] The data acquisition unit is used to collect various parameters of the decentralized building heating system, including heating equipment power parameters, pipe thermal conductivity parameters, and indoor and outdoor temperature parameters.

[0112] The feature extraction unit is connected to the data acquisition unit and performs multi-scale feature extraction on the heating area based on the improved SSD algorithm to obtain a feature data set.

[0113] The time series and population analysis unit is connected to the feature extraction unit. Using the heating time series and population optimization model, it establishes the correlation mapping relationship between heating time series and population distribution to obtain dynamic data sequence of population distribution.

[0114] The optimization decision unit is connected to the feature extraction unit, the time series and population analysis unit, and the data acquisition unit, respectively. It performs calculations based on the input data and outputs an optimized combination scheme of heating parameters.

[0115] The parameter adjustment unit is connected to the optimization decision unit and adjusts the heating system parameters in real time according to the optimization combination scheme.

[0116] The feedback update unit is connected to the parameter adjustment unit and the optimization decision unit, and feeds back the adjusted heating system operation data to the optimization decision unit to iteratively update the optimization decision model.

[0117] The intelligent optimization method and system for distributed building heating parameters utilizes an improved SSD algorithm to extract multi-scale features from the heating area, capturing environmental information across different spatial dimensions. Simultaneously, it employs a heating time-series and population optimization model to deeply correlate historical heating data with population flow data, establishing a precise time-series-population distribution mapping relationship. This data-driven analysis model enables the system to perceive real-time changes in indoor and outdoor temperature fluctuations, population gathering or evacuation, etc., providing comprehensive and dynamic evidence for parameter optimization and completely eliminating reliance on empirical rules.

[0118] At the algorithmic model level, traditional technologies, due to their singular and outdated models, fail to uncover the potential relationships between parameters, leading to inefficient heating regulation. This system's optimization decision-making model integrates a multi-layered neural network and innovatively introduces several formulas. The heating parameter optimization formula comprehensively considers the feature data processed by the improved SSD algorithm, the dynamic data output from the heating time series and population optimization model, and the standardized heating parameter vector, outputting a precise parameter combination scheme through weighted calculation. The dynamic adjustment formula combines the time rate of change, the population flow influence coefficient, and parameter coupling relationships to achieve real-time dynamic adjustment of parameters. These formulas work together to enable the system to deeply analyze the complex characteristics of the heating system, accurately grasp the inherent logic between parameters, and significantly improve the scientific rigor and effectiveness of optimization decisions.

[0119] The system further enhances its advantages through a closed-loop operation mechanism. The data acquisition unit accurately obtains parameters such as heating equipment power and pipe thermal conductivity coefficients, which are then normalized and input into the optimization decision-making unit. The parameter adjustment unit adjusts the heating system in real time according to the output optimization plan and feeds back the adjusted operating data to the optimization decision-making unit for iterative updates. This continuous optimization model not only ensures the stability and comfort of the heating effect but also significantly improves energy efficiency, effectively avoiding the energy waste caused by the crude parameter adjustments in traditional technologies. It provides an efficient and intelligent innovative solution for decentralized building heating.

[0120] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0121] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent optimization of heating parameters in decentralized buildings, characterized in that, Includes the following steps: Step S1: Based on the improved SSD algorithm, perform multi-scale feature extraction on the heating area of ​​the distributed building to obtain the feature data set of the heating area under different spatial dimensions; Step S2: Using the heating time series and population optimization model, combined with historical heating data and historical population flow data within buildings, establish the correlation mapping relationship between heating time series and population distribution to obtain a dynamic data sequence of population distribution changes over time. Step S3: Collect various parameters of the decentralized building heating system, including the power parameters of the heating equipment, the thermal conductivity parameters of the pipes, and the indoor and outdoor temperature parameters. Normalize the collected parameters to form a standardized parameter vector. Step S4: Input the feature data set, dynamic data sequence and standardized parameter vector into the optimization decision model, and output the optimized combination scheme of heating parameters through the multi-layer neural network calculation of the optimization decision model; Step S5: Based on the optimized combination scheme, adjust the parameters of the decentralized building heating system in real time and record the adjusted heating system operation data simultaneously; Step S6: Feed the running data back to the optimization decision model at set time intervals, iterate and update the optimization decision model, and continuously optimize the combination scheme of heating parameters. The optimization decision model includes heating parameter optimization formulas: in, The optimized heating parameter combination scheme; This is the weighting adjustment coefficient; This refers to the number of data samples. Indicates the first Feature data of each heating area extracted using the improved SSD algorithm; To improve the SSD algorithm for feature data The processing function; Indicates the first Dynamic data on the distribution of the population within a building over a given time period; For the heating time series and population optimization model, population dynamic data The processing function; Indicates the first A standardized heating parameter vector at each sampling time; For standardized parameter vectors Parameter processing functions; In step S4, the optimization decision model further includes a dynamic adjustment formula for heating parameters: in, This refers to the adjustment amount of heating parameters; This is the time sensitivity coefficient; To optimize the solution Rate of change over time; The population mobility impact coefficient reflects the degree of influence of population distribution changes on heating parameters; This is the comprehensive influence coefficient of heating parameters, reflecting the coupling effect between various heating parameters; In step S1, when the improved SSD algorithm extracts features from the heating area, it employs a multi-scale sliding window mechanism and calculates the feature response values ​​at different scales using the following formula: in, The feature response value at a preset scale; The number of feature dimensions; For the first Weights of each feature dimension; To analyze the characteristic data of heating areas at a preset scale Transformation function; In step S2, the heating timing and population optimization model constructs a population distribution prediction formula: in, For the future Population distribution projections for a given time period; Select the length for historical data; Historical population data Weighting coefficients; This is the error term for population distribution prediction; In step S3, the collected heating parameters are normalized using the following formula: in, These are the normalized parameter values; These are the original parameter values; This is the maximum value of the parameter in historical data; This is the minimum value of the parameter in historical data.

2. The intelligent optimization method for distributed building heating parameters according to claim 1, characterized in that, Step S3 includes the following sub-steps: Step S3.1: Use a high-precision sensor array to collect the power parameters of the heating equipment in real time; Step S3.2: Using a thermal conductivity coefficient detection device, the thermal conductivity coefficient parameter of the pipeline is measured based on the principle of heat flow meter to obtain the thermal conductivity performance data of the pipeline under different environmental conditions; Step S3.3: By using temperature sensors placed inside and outside the building, indoor and outdoor temperature parameters are periodically collected according to the set sampling frequency to form a continuous temperature data sequence; Step S3.4: The collected power parameters of heating equipment, thermal conductivity parameters of pipes, and indoor and outdoor temperature parameters are processed to unify the data format and convert them into a standard data format that meets the system requirements.

3. The intelligent optimization method for decentralized building heating parameters according to claim 1, characterized in that, Step S4 includes the following sub-steps: Step S4.1: Perform data fusion processing on the feature data set, dynamic data sequence, and standardized parameter vector, and construct a complete input data matrix through feature concatenation and temporal alignment techniques; Step S4.2: Input the input data matrix into the input layer of the optimization decision model. The data is processed by the convolutional layer in the model to extract features and reduce dimensionality, thereby enhancing the expressive power of the data features. Step S4.3: By optimizing the recurrent neural network layer of the decision model, the temporal characteristics of the data are learned and memorized to capture the pattern of heating parameters changing over time; Step S4.4: The output layer of the optimization decision model calculates the processed data, outputs an optimized combination scheme of heating parameters, and performs data format conversion to meet the execution requirements of the heating system.

4. The intelligent optimization method for distributed building heating parameters according to claim 1, characterized in that, Step S5 includes the following sub-steps: Step S5.1: Based on the optimized combination scheme, generate the corresponding parameter adjustment instruction set, and clarify the adjustment target value and adjustment sequence of each heating equipment parameter; Step S5.2: Send the parameter adjustment command set to the controller of the heating system through the communication module; Step S5.3: The heating system controller adjusts the parameters of the heating equipment parameter by parameter according to the received instruction set, and monitors the equipment operating status in real time during the adjustment process; Step S5.4: After the parameters are adjusted, start the data recording module to collect and store the adjusted heating system operation data in all aspects, including equipment operation parameters and indoor and outdoor environmental parameters.

5. A decentralized building heating parameter intelligent optimization system, characterized in that, This system is applied to the intelligent optimization method for distributed building heating parameters as described in claim 1, comprising: The data acquisition unit is used to collect various parameters of the decentralized building heating system, including heating equipment power parameters, pipe thermal conductivity parameters, and indoor and outdoor temperature parameters. The feature extraction unit is connected to the data acquisition unit and performs multi-scale feature extraction on the heating area based on the improved SSD algorithm to obtain a feature data set. The time series and population analysis unit is connected to the feature extraction unit. Using the heating time series and population optimization model, it establishes the correlation mapping relationship between heating time series and population distribution to obtain dynamic data sequence of population distribution. The optimization decision unit is connected to the feature extraction unit, the time series and population analysis unit, and the data acquisition unit, respectively. It performs calculations based on the input data and outputs an optimized combination scheme of heating parameters. The parameter adjustment unit is connected to the optimization decision unit and adjusts the heating system parameters in real time according to the optimization combination scheme. The feedback update unit is connected to the parameter adjustment unit and the optimization decision unit, and feeds back the adjusted heating system operation data to the optimization decision unit to iteratively update the optimization decision model.

Images