Building overall performance control method and system

Abstract

The invention discloses a building overall performance control method and system. The method comprises the following steps: determining a corresponding comprehensive weight according to an electric equipment low-carbon evaluation model and multiple training of a plurality of weights of different dimensions; the real-time performance score of each electric device is determined according to the comprehensive weight and the corresponding electric device, a low-carbon evaluation model of the electric device and multiple training of the weights of multiple different dimensions are compatible, the accuracy of the comprehensive weight is ensured, multiple operations of the weights of multiple different dimensions are fully utilized, and the accuracy of the comprehensive weight is improved. And control from multiple dimensions is realized. Determining the overall performance score of the building based on the real-time performance score of each electric device and the power utilization weight of each electric device; and dynamically optimizing the operation strategy of each electric device according to the overall performance score of the building, presenting the overall performance effect of the building based on the overall performance score of the building, and dynamically optimizing the operation strategy of each electric device.

Description

Technical Field

[0001] The present invention relates to the technical field of overall performance control methods, and particularly to a method and system for controlling the overall performance of a building. Background Art

[0002] With the development of technology, there are various concepts of building energy conservation in existing designs, including two-star green buildings, three-star green buildings, ultra-low energy consumption, near-zero energy consumption, zero energy consumption, low carbon, near-zero carbon, and zero carbon. These belong to different building energy conservation standards, and their focus on energy conservation points is different. Moreover, the parameters among them overlap, and it is necessary to combine and evaluate through the parameter values of different systems in the building. With the strong promotion of the digital transformation of buildings, various digital platforms emerge in an endless stream, and the monitored data is mostly primary data of equipment sensors. However, these data are only simply presented or presented after simple calculations, without being fused and analyzed, and cannot be controlled from multiple dimensions, affecting the good effect of the overall performance of the building. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a method for controlling the overall performance of a building, which specifically includes: Determine the operation data set of each electrical device according to the building operation and maintenance platform and each electrical device in the building; Determine the corresponding low-carbon evaluation model for the electrical device based on the operation data set of each electrical device; Determine the corresponding comprehensive weight according to the low-carbon evaluation model of the electrical device and multiple trainings with weights of multiple different dimensions; Determine the real-time performance score of each electrical device according to the comprehensive weight and the corresponding electrical device; Determine the overall performance score of the building based on the real-time performance scores of each electrical device and the electricity consumption weights of each electrical device; Dynamically optimize the operation strategy of each electrical device according to the overall performance score of the building.

[0004] Preferably, the step of determining the operation data set of each electrical device according to the building operation and maintenance platform and each electrical device in the building includes: Collect the building name and determine the building operation and maintenance platform according to the building name and the building database; Determine each electrical device in the building based on traversing the building database; Interact with the operation and maintenance platform and each electrical device in the building, and determine the operation data set of each electrical device according to the interaction between the building operation and maintenance platform and each electrical device in the building.

[0005] Preferably, determining the corresponding low-carbon evaluation model for the electrical equipment based on the operation data sets of each electrical equipment includes: Collect the operation data sets of each electrical equipment, and determine multiple operation data combinations under different operation types based on the detection of the operation data sets of each electrical equipment; Determine the corresponding low-carbon evaluation model for the electrical equipment according to the multiple trainings of the multiple operation data combinations, output the low-carbon evaluation results of the corresponding electrical equipment according to the low-carbon evaluation model of the electrical equipment and the corresponding electrical equipment, and dynamically adjust the operation strategy of the chiller according to the low-carbon evaluation results of each electrical equipment.

[0006] Preferably, determining the corresponding comprehensive weight according to the multiple trainings of the low-carbon evaluation model of the electrical equipment and the weights of multiple different dimensions includes: In the low-carbon evaluation model of the electrical equipment, construct an initial judgment matrix based on the past operation data of each electrical equipment, and determine the initial weight of the network analysis ANP of each influencing factor according to the initial judgment matrix ; Calculate the information entropy of each influencing factor, and trigger the correction of the initial weight of each influencing factor according to the information entropy of each influencing factor to output the corrected weight by the entropy weight method ; Train the long short-term memory network LSTM model based on the past operation data of each electrical equipment, perform transfer learning on the real-time data of the building's operation and maintenance platform, optimize its prediction accuracy, use the mean square error MSE as the loss function, and perform training through the optimizer Adam to output the prediction result of the LSTM model; Dynamically correct the weights of each influencing factor according to the prediction result of the LSTM model to output the corrected weight of the LSTM .

[0007] Preferably, determining the corresponding comprehensive weight according to the multiple trainings of the low-carbon evaluation model of the electrical equipment and the weights of multiple different dimensions further includes: Take the ANP initial weight , the corrected weight by the entropy weight method and the corrected weight of the LSTM as inputs, and calculate the attention scores of various weights according to the attention mechanism model; according to the attention scores of various weights, perform weighted fusion on the three weights to obtain the comprehensive weight .

[0008] Preferably, determining the real-time performance score of each electrical equipment according to the comprehensive weight and the corresponding electrical equipment includes: Determine the real-time performance scores of each electrical equipment in the building based on the comprehensive weight and the corresponding electrical equipment ; Real-time performance scores of each electrical device The calculation formula is as follows: .

[0009] Preferably, determining the overall performance score of the building based on the real-time performance scores of each electrical device and the electrical weights of each electrical device includes: Collecting the real-time power consumption data of each electrical device, and determining the electrical weights of each electrical device for the overall performance of the building according to the real-time power consumption data of each electrical device ; According to the real-time performance scores of each electrical device and the electrical weights Calculate the overall performance score of the building.

[0010] Preferably, determining the overall performance score of the building based on the real-time performance scores of each electrical device and the electrical weights of each electrical device further includes: The electrical weights of each electrical device for the overall performance of the building The calculation formula is as follows: ; Where: is the real-time power consumption of the electrical device ; is the total number of electrical devices in the building; The calculation formula for the overall performance score of the building is as follows: ; Where: S is the attenuation coefficient of the building envelope, S which decays with the age of the building.

[0011] Preferably, dynamically optimizing the operation strategies of each electrical device according to the overall performance score of the building includes: Dividing the overall performance of the building into multiple levels according to the overall performance score of the building; Dynamically optimizing the operation strategies of each electrical device according to the real-time performance scores of each electrical device and the overall performance of the building.

[0012] The present invention also discloses a building overall performance control system. The building overall performance control system is applied to the building overall performance control method as described in any one of the foregoing, and the building overall performance control system includes: A data set module for determining the operation data set of each electrical device according to the building operation and maintenance platform and each electrical device in the building; The low-carbon evaluation module for electrical equipment is used to determine the corresponding low-carbon evaluation model for electrical equipment based on the operation data set of each electrical equipment; The comprehensive weight module is used to determine the corresponding comprehensive weight according to the multiple training of the low-carbon evaluation model of electrical equipment and the weights of multiple different dimensions; The real-time performance scoring module for electrical equipment is used to determine the real-time performance score of each electrical equipment according to the comprehensive weight and the corresponding electrical equipment; The overall performance scoring module is used to determine the overall performance score of the building based on the real-time performance scores of each electrical equipment and the electricity consumption weights of each electrical equipment; The optimization module is used to dynamically optimize the operation strategies of each electrical equipment according to the overall performance score of the building.

[0013] The present invention has the following beneficial effects: (1) Determine the operation data set of each electrical equipment according to the operation and maintenance platform of the building and each electrical equipment in the building; determine the corresponding low-carbon evaluation model for electrical equipment based on the operation data set of each electrical equipment; determine the corresponding comprehensive weight according to the multiple training of the low-carbon evaluation model of electrical equipment and the weights of multiple different dimensions; determine the real-time performance score of each electrical equipment according to the comprehensive weight and the corresponding electrical equipment, which is compatible with the multiple training of the low-carbon evaluation model of electrical equipment and the weights of multiple different dimensions, ensures the accuracy of the comprehensive weight, and makes full use of the multiple operations of the weights of multiple different dimensions to achieve control from multiple dimensions.

[0014] (2) Determine the overall performance score of the building based on the real-time performance scores of each electrical equipment and the electricity consumption weights of each electrical equipment; dynamically optimize the operation strategies of each electrical equipment according to the overall performance score of the building, present the overall performance effect of the building based on the overall performance score of the building, and dynamically optimize the operation strategies of each electrical equipment, ensuring the good operation effect of each electrical equipment and the good effect of the overall performance of the building. Description of the Drawings

[0015] Figure 1 It is a schematic diagram of the application scenario of the overall performance control method for a building in an embodiment.

[0016] Figure 2 It is a schematic flowchart of the overall performance control method for a building in an embodiment of the present invention.

[0017] Figure 3 It is a schematic diagram of the structural composition of the overall performance control system for a building in an embodiment of the present invention. Detailed Embodiments

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0019] The overall performance control method for a building provided in this application is applied to an application environment as Figure 1 shown. Among them, the computer 102 communicates with the server 104 through a network. Among them, the computer 102 is but not limited to various personal computers, servers, and overall performance control methods, and the server 104 is implemented by an independent server or a server cluster composed of servers.

[0020] Please refer to Figures 1 to 3 , an overall performance control method for a building, which is applied to the overall performance control scenario of a building; the overall performance control method for a building includes: Step S11: Determine the operation data set of each electrical device according to the operation and maintenance platform of the building and each electrical device in the building.

[0021] Step S12: Determine the corresponding low-carbon evaluation model for the electrical device based on the operation data set of each electrical device.

[0022] Step S13: Determine the corresponding comprehensive weight according to the low-carbon evaluation model of the electrical device and multiple trainings of weights in multiple different dimensions.

[0023] Step S14: Determine the real-time performance score of each electrical device according to the comprehensive weight and the corresponding electrical device.

[0024] Step S15: Determine the overall performance score of the building based on the real-time performance scores of each electrical device and the electricity consumption weights of each electrical device.

[0025] Step S16: Dynamically optimize the operation strategy of each electrical device according to the overall performance score of the building.

[0026] In step S11, the operation data set of each electrical device is determined according to the operation and maintenance platform of the building and each electrical device in the building.

[0027] In the specific implementation process of the present invention, the specific steps are: Step S111: Determine each electrical device in the building based on traversing the building database.

[0028] Step S112: Interact with the operation and maintenance platform of the building and each electrical device in the building, and determine the operation data set of each electrical device according to the interaction between the operation and maintenance platform of the building and each electrical device in the building.

[0029] In the embodiments of the present application, each electrical device in the building is determined based on the traversal of the building database, ensuring subsequent management and control of each electrical device in the building.

[0030] At this time, the building database is introduced and traversed, so as to output each electrical device in the building during the traversal process, facilitating the introduction of each electrical device in the building, and then subsequent management and control of each electrical device in the building.

[0031] Furthermore, an interaction is carried out between the building operation and maintenance platform and each electrical device in the building. Based on the interaction between the building operation and maintenance platform and each electrical device in the building, a set of operation data of each electrical device is determined. The set of operation data of each electrical device is introduced to manage and control each electrical device.

[0032] At this time, the building operation and maintenance platform and each electrical device in the building are introduced. The building operation and maintenance platform interacts with each electrical device in the building to monitor the operation of each electrical device and outputs a set of operation data of each electrical device.

[0033] At the same time, real-time operation data of each type of electrical device is collected from the building operation and maintenance platform, including: Performance data: such as the energy efficiency of the chiller, the brightness of the lighting system, etc. Electrical consumption data: such as the real-time power and energy consumption of the electrical device. Building nature: such as hospital, school, shopping mall, airport, high-speed railway station, etc.

[0034] At this time, taking the chiller as an example for the real-time operation data of the electrical device: Energy efficiency data: cooling capacity, energy consumption, coefficient of performance (COP), etc. Reliability data: failure rate, mean time between failures (MTBF), maintenance frequency, etc. Environmental data: noise, refrigerant leakage, carbon emissions, etc. Economic data: operation cost, maintenance cost, downtime loss, etc.

[0035] At the same time, the data is processed such as cleaning, deduplication, and normalization to ensure data quality. The data is organized in a time series, which is convenient for subsequent analysis. The data is organized in a time series, which is convenient for the input of the LSTM model and forms a set of operation data of each electrical device.

[0036] In step S12, a corresponding low-carbon evaluation model for the electrical device is determined based on the set of operation data of each electrical device. In the specific implementation process of the present invention, the specific steps are as follows: Step S121: Collect the set of operation data of each electrical device, and determine multiple operation data combinations under different operation types based on the detection of the set of operation data of each electrical device.

[0037] Step S122: Determine the corresponding low-carbon evaluation model of the electrical equipment according to the multiple training of multiple operation data combinations, output the low-carbon evaluation result of the corresponding electrical equipment according to the low-carbon evaluation model of the electrical equipment and the corresponding electrical equipment, and dynamically adjust the operation strategy of the chiller according to the low-carbon evaluation results of each electrical equipment.

[0038] In the embodiment of the present application, the operation data sets of each electrical equipment are collected, and multiple operation data combinations under different operation types are determined based on the detection of the operation data sets of each electrical equipment. By controlling different operation types, multiple operation data combinations under different operation types are introduced.

[0039] At this time, the operation data sets of each electrical equipment are collected, and the operation data sets of each electrical equipment are detected, so as to determine multiple operation data combinations under different operation types based on the detection of the operation data sets of each electrical equipment, realizing the internal classification of the operation data sets of each electrical equipment, facilitating subsequent control for different operation types, and introducing multiple operation data combinations under different operation types.

[0040] Therefore, determine the corresponding low-carbon evaluation model of the electrical equipment according to the multiple training of multiple operation data combinations, output the low-carbon evaluation result of the corresponding electrical equipment according to the low-carbon evaluation model of the electrical equipment and the corresponding electrical equipment, and dynamically adjust the operation strategy of the chiller according to the low-carbon evaluation results of each electrical equipment, ensuring the accuracy of the low-carbon evaluation results of each electrical equipment and further dynamically adjusting the operation strategy of the chiller.

[0041] At this time, multiple operation data combinations are introduced, controlled, and the corresponding low-carbon evaluation model of the electrical equipment is determined according to the multiple training of multiple operation data combinations, thereby utilizing multiple operation data combinations, performing multiple training based on multiple operation data combinations and the corresponding electrical equipment, and gradually forming the corresponding low-carbon evaluation model of the electrical equipment.

[0042] Meanwhile, output the low-carbon evaluation result of the corresponding electrical equipment according to the low-carbon evaluation model of the electrical equipment and the corresponding electrical equipment, introduce the low-carbon evaluation result of the corresponding electrical equipment, and dynamically adjust the operation strategy of the chiller according to the low-carbon evaluation results of each electrical equipment, thereby controlling the operation strategy of the chiller.

[0043] Optionally, considering the reasonable decline in the performance of the electrical equipment in the building over time, establish the low-carbon evaluation model of the j-th type of electrical equipment in the building as follows: .

[0044] In the formula, 、 , …… is the fusion weight of the k-th influence parameter corresponding to the j-th electrical equipment; P j Incrementing according to the number of startups, when the building is initially delivered for the first time (P = 0), the j performance A of the electrical equipment j = 100%; Based on the A j scoring value, dynamically adjust the operation strategy of the chiller, such as frequency conversion adjustment, load distribution, etc.

[0045] In step S13, according to the low-carbon evaluation model of electrical equipment and multiple trainings of weights in different dimensions, determine the corresponding comprehensive weight.

[0046] In the specific implementation process of the present invention, the specific steps are as follows: Step S131: In the low-carbon evaluation model of electrical equipment, construct an initial judgment matrix based on the past operation data of each electrical equipment, and determine the ANP initial weight of each influencing factor according to the initial judgment matrix .

[0047] Step S132: Calculate the information entropy of each influencing factor, and trigger the correction of the initial weight of each influencing factor according to the information entropy of each influencing factor to output the weight corrected by the entropy weight method .

[0048] Step S133: Train the LSTM model based on the past operation data of each electrical equipment, perform transfer learning on the real-time data of the building's operation and maintenance platform, optimize its prediction accuracy, and use the mean square error MSE as the loss function and the Adam optimizer for training to output the prediction result of the LSTM model.

[0049] Step S134: Dynamically correct the weights of each influencing factor according to the prediction result of the LSTM model to output the LSTM corrected weight .

[0050] Step S135: Use the ANP initial weight , the weight corrected by the entropy weight method and the LSTM corrected weight as inputs, and calculate the attention scores of various weights according to the attention mechanism model; According to the attention scores of various weights, perform weighted fusion on the three weights to obtain the comprehensive weight .

[0051] In the embodiment of the present application, in the low-carbon evaluation model of electrical equipment, construct an initial judgment matrix based on the past operation data of each electrical equipment, and determine the ANP initial weight of each influencing factor according to the initial judgment matrix , ensuring the accuracy of the ANP initial weights of each influencing factor accuracy.

[0052] At this time, the past operation data of each electrical equipment is introduced, and the past operation data of each electrical equipment is controlled to construct an initial judgment matrix for further control of the initial judgment matrix. At the same time, the ANP initial weights of each influencing factor are determined according to the initial judgment matrix , ensuring the accuracy of the ANP initial weights of each influencing factor accuracy.

[0053] Furthermore, the information entropy of each influencing factor is calculated, and the initial weights of each influencing factor are corrected according to the information entropy of each influencing factor to output the entropy weight method corrected weights , and the entropy weight method corrected weights are further introduced .

[0054] At this time, according to the multi-source data, the information entropy of each influencing factor is calculated ; ; wherein The probability of the jth influencing factor in the ith sample.

[0055] Entropy weight method corrected weights : .

[0056] Furthermore, based on the past operation data of each electrical equipment, an LSTM model is trained to perform transfer learning on the real-time data of the building's operation and maintenance platform, optimize its prediction accuracy, and use the mean square error MSE as the loss function and the Adam optimizer for training to output the prediction results of the LSTM model, and the prediction results of the LSTM model are introduced.

[0057] At this time, an LSTM model is trained based on the past operation data of each electrical equipment, and the LSTM model is introduced. In the LSTM model, input: historical operation data such as energy consumption and failure rate in the past 24 hours. Output: predicted values of energy consumption, failure rate, etc. in the next period such as the next hour.

[0058] At the same time, a multi-layer LSTM network is used to capture the long-term dependence relationship of time series data. A Dropout layer is added to prevent overfitting. The LSTM model is trained using historical data to perform transfer learning on the real-time data of multiple operation and maintenance platforms and optimize its prediction accuracy.

[0059] The mean square error MSE is used as the loss function and the Adam optimizer is used for training; according to the prediction results of the LSTM, the weights of each influencing factor are dynamically corrected For example, if the LSTM predicts that the energy consumption will increase significantly, the weight of energy consumption is increased. If the LSTM predicts that the failure rate will decrease, the weight of the failure rate is decreased.

[0060] Furthermore, the weights of various influencing factors are dynamically corrected according to the prediction results of the LSTM model to output the LSTM corrected weights , ensuring the accuracy of the LSTM corrected weights .

[0061] At this time, the prediction results of the LSTM model are introduced to dynamically correct the weights of various influencing factors. At the same time, the LSTM corrected weights are output , ensuring the accuracy of the LSTM corrected weights .

[0062] Therefore, the ANP initial weights , the entropy weight method corrected weights and the LSTM corrected weights are used as inputs, and the attention scores of various weights are calculated according to the attention mechanism model; according to the attention scores of various weights, the three weights are weighted and fused to obtain the comprehensive weight .

[0063] At this time, the ANP initial weights , the entropy weight method corrected weights and the LSTM corrected weights are introduced, and the ANP initial weights , the entropy weight method corrected weights and the LSTM corrected weights are weighted and fused to obtain the comprehensive weight .

[0064] Design an attention mechanism model to calculate the attention scores of the three weights. Formula: ; where is the importance function of the weight.

[0065] Fusion weight: According to the attention scores, the three weights are weighted and fused to obtain the comprehensive weight : .

[0066] In step S14, the real-time performance scores of each electrical device are determined according to the comprehensive weight and the corresponding electrical device; in the specific implementation process of the present invention, the specific steps are: Step S141: Determine the real-time performance scores of each electrical device in the building based on the comprehensive weight and the corresponding electrical device .

[0067] In an embodiment of the present application, based on the comprehensive weight and the corresponding electrical equipment, the real-time performance scores of each electrical equipment in the building are determined, which is compatible with the multiple training of the low-carbon evaluation model of the electrical equipment and the weights of multiple different dimensions, ensuring the accuracy of the comprehensive weight and making full use of the multiple operations of the weights of multiple different dimensions to achieve control from multiple dimensions.

[0068] The real-time performance scores of each electrical equipment The calculation formula is as follows: 。

[0069] According to the real-time performance scores of each electrical equipment the operation strategies of the electrical equipment are adjusted, such as frequency conversion regulation, load distribution, etc.

[0070] In step S15, based on the real-time performance scores of each electrical equipment and the electrical weights of each electrical equipment, the overall performance score of the building is determined; In the specific implementation process of the present invention, the specific steps are as follows: Step S151: Collect the real-time electricity consumption data of each electrical equipment, and determine the electrical weight of each electrical equipment for the overall performance of the building according to the real-time electricity consumption data of each electrical equipment 。

[0071] Step S152: Calculate the overall performance score of the building according to the real-time performance scores and electrical weights of each electrical equipment.

[0072] The electrical weight of each electrical equipment for the overall performance of the building The calculation formula is as follows: ; Where: is the real-time electricity consumption of the electrical equipment ; is the total number of electrical equipment in the building; The calculation formula for the overall performance score of the building is as follows: ; where S is the attenuation coefficient of the building envelope, and S decays with the age of the building.

[0073] In an embodiment of the present application, the real-time electricity consumption data of each electrical equipment is collected, and the electrical weight of each electrical equipment for the overall performance of the building is determined according to the real-time electricity consumption data of each electrical equipment ensuring the accuracy of the electrical weight of each electrical equipment for the overall performance of the building.

[0074] In step S16, the operation strategies of each electrical device are dynamically optimized according to the overall performance score of the building; in the specific implementation process of the present invention, the specific steps are as follows: Step S161: Divide the overall performance of the building into multiple levels according to the overall performance score of the building.

[0075] Step S162: Dynamically optimize the operation strategies of each electrical device according to the real-time performance score of each electrical device and the overall performance of the building.

[0076] In the embodiment of the present application, dividing the overall performance of the building into multiple levels according to the overall performance score of the building realizes the division of the overall performance of the building, facilitating subsequent control of multiple levels.

[0077] Therefore, dynamically optimizing the operation strategies of each electrical device according to the real-time performance score of each electrical device and the overall performance of the building, presenting the overall performance effect of the building based on the overall performance score of the building, and dynamically optimizing the operation strategies of each electrical device ensure the good operation effect of each electrical device and the good effect of the overall performance of the building.

[0078] Optionally, divide the overall performance of the building into different levels such as excellent, good, average, and poor according to the comprehensive score LCB.

[0079] Based on the score of each electrical device, give suggestions, such as optimizing the operation strategy of electrical devices with high energy consumption, adjusting the usage time of electrical devices, and reducing the peak load; when < 80, it is recommended to check the electrical device. When < 70, it is recommended to maintain the electrical device. When < 60, it is recommended to replace the parts of the electrical device. When < 40, it is recommended to repair or replace the electrical device with poor performance. The above scores are adjusted according to each electrical device.

[0080] Specifically, in a large commercial building, the method of the present invention is applied to evaluate and optimize the operation performance of the building in real time.

[0081] Implementation steps: Collect the real-time operation data and power consumption data of each device from the building operation and maintenance platform. Build an LSTM model to predict the future change trends of various influencing factors. Use the LSTM prediction results to dynamically correct the weights of various influencing factors. Further correct the weights using the entropy weight method to enhance objectivity. Through the attention mechanism, fuse the ANP initial weight, the LSTM corrected weight, and the entropy weight method corrected weight to obtain the comprehensive weight. Based on the comprehensive weight, calculate the real-time performance scores of various devices: Calculate the electricity consumption weight based on the real-time electricity consumption data of the equipment: Optimize the operation strategy of the j-th type of equipment according to the performance score. Display the real-time performance score and optimization suggestions through the user interface.

[0082] Calculate the overall performance score of the building based on the equipment performance score and electricity consumption weight; Generate building efficiency improvement suggestions according to the evaluation results, such as: Repair or replace equipment with poor performance, and optimize the operation strategy of equipment with high energy consumption. Display the overall performance score and efficiency improvement suggestions of the building through the user interface.

[0083] Implementation effect: Improved the overall energy efficiency of the building by 12%, reduced energy consumption. Reduced the failure rate and improved reliability. Optimized the operation cost and reduced the economic expenditure.

[0084] In the airport application scenario, airport buildings usually have large spaces, high passenger flow densities, and complex equipment systems such as chillers, lighting systems, elevators, etc. The operation performance of the building equipment directly affects energy consumption and passenger comfort.

[0085] Implementation steps: Collect multi-source data: Equipment operation data: Chiller energy consumption, lighting system brightness, elevator operation times, etc. Environmental data: Outdoor temperature, humidity, wind speed, etc. Passenger flow data: Passenger flow, flight information, etc. Collect real-time data: Collect data in real-time through sensors and monitoring systems.

[0086] Efficiency improvement suggestions: During peak hours such as peak flight hours, optimize the operation strategies of chillers and lighting systems to improve energy efficiency. During off-peak hours such as at night, reduce the operation load of chillers and lighting systems to reduce energy consumption.

[0087] Implementation effect: Improved the energy efficiency of airport buildings and reduced energy consumption. Improved passenger comfort and reduced complaints.

[0088] In the school application scenario, school buildings usually have periodic usage characteristics such as class hours and holidays. The operation performance of the buildings directly affects energy consumption and student comfort.

[0089] Implementation steps: Collect multi-source data: Chiller energy consumption, lighting system brightness, operation times of multimedia equipment, etc. Environmental data: Outdoor temperature, humidity, wind speed, etc. Usage data: Class schedules, classroom usage, etc. Collect real-time data: Collect data in real-time through sensors and monitoring systems.

[0090] Efficiency improvement suggestions: During class hours, optimize the operation strategies of chillers and lighting systems to improve energy efficiency. During holidays or non-class hours, reduce the operation load of chillers and lighting systems to reduce energy consumption.

[0091] Implementation effect: The energy efficiency of school buildings has been increased by 21%, and energy consumption has been reduced. The comfort of students has been improved, and the learning environment has been enhanced.

[0092] Buildings are often designed in detail from different perspectives during the design phase. However, in actual use, they often do not operate as envisioned. For three-star buildings, zero-energy buildings, or zero-carbon buildings designed, there is no appropriate method to evaluate whether they can truly achieve the expected energy savings. Moreover, the requirements in design specifications are mostly for the virtual annual energy consumption, and there is no evaluation of the low-carbon effect under a certain period of time, a certain type of weather, a certain season, or a certain utilization rate of the building.

[0093] Through the method of the present invention, only a small amount of incomplete real-time building data training is required to achieve the assessment of the building's energy-saving situation, thereby improving the efficiency and practicality of the assessment, and filling the gap in the lack of immediate evaluation based on actual data during the building operation and maintenance stage. With the advancement of building digitization, this method can be more conveniently implemented and provide a real and effective database for various types of buildings.

[0094] In addition, based on the factory data of equipment, different specifications, and data from different building platforms, a low-carbon evaluation model for equipment is constructed. The ANP method, LSTM, and entropy weight method combined with the attention mechanism are used to triple-correct the coefficients. This improvement enhances the model's attention to key information, accelerates the model's convergence process, successfully reduces the number of training times, and simultaneously improves the accuracy and reliability of the model results. Through the progressive weight optimization process of ANP (initial weight) → entropy weight method (objective correction) → LSTM (dynamic adjustment) → attention mechanism (fusion weight), a closed-loop feedback is formed, significantly enhancing the scientific nature of the evaluation.

[0095] In multi-source data fusion, the real-time operation data, environmental data (temperature, humidity, wind speed), and usage data (number of people, schedule) of multiple devices in the building, such as chiller hosts, lighting systems, and elevators, are integrated, breaking through the limitations of traditional single-data-source evaluation. Traditional methods usually rely on a single data source and may have biases. Multi-source data fusion can significantly improve the accuracy and reliability of the evaluation.

[0096] In the dynamic weight adjustment mechanism, LSTM transfer learning is combined to predict time-series data and dynamically correct the weights, solving the defect that traditional static models such as ANP cannot adapt to real-time changes. Traditional methods usually use static weights and are difficult to adapt to the changes in real-time data. Dynamic weight adjustment can respond more flexibly to changes in the operating environment.

[0097] In the entropy weight method and the attention mechanism, based on the information entropy of multi-source data, the LSTM prediction results, the weights corrected by the entropy weight method, and the ANP initial weights are fused through the attention mechanism to obtain a more scientific comprehensive weight. The entropy weight method enhances the objectivity of the weights, and the attention mechanism can more scientifically fuse weights from different sources, improving the scientificity and accuracy of the evaluation.

[0098] In real-time performance evaluation and optimization, based on the fused weights, the overall performance score of the building is calculated in real-time, and targeted efficiency improvement suggestions are generated. Traditional methods usually can only provide static evaluation results and are difficult to achieve real-time optimization. Real-time performance evaluation and optimization can significantly improve the energy efficiency and operation efficiency of buildings.

[0099] In comprehensively considering various factors, various factors such as the performance of equipment, electricity consumption data, outdoor weather, and seasonal changes are comprehensively considered to provide a more comprehensive performance evaluation. Traditional methods usually only consider a single or a few factors and are difficult to comprehensively reflect the operating state of the building. Comprehensively considering various factors can provide a more comprehensive performance evaluation.

[0100] Using electricity consumption data as weights for the overall performance evaluation of buildings: Combining the real-time electricity consumption data of equipment, calculating the contribution weights of each equipment to the overall performance of the building, and comprehensively evaluating the overall performance of the building without relying on manual experience to assign weights. Taking the real-time electricity consumption of equipment as weights, constructing an overall performance scoring model of the building to achieve two-dimensional optimization of "equipment performance - energy consumption contribution".

[0101] The present invention proposes a brand-new method for building performance evaluation through multi-source data fusion, dynamic weight adjustment, and the combination of the entropy weight method and the attention mechanism. The present invention not only integrates a variety of technologies but also realizes the dynamic correction and scientific fusion of weights through LSTM network transfer learning and the attention mechanism.

[0102] Through dynamic weight adjustment and real-time data prediction, the present invention can respond to changes in the operating environment in real-time, significantly improving the accuracy and reliability of the evaluation. Through multi-source data fusion and comprehensively considering various factors, the present invention can provide a more comprehensive and scientific performance evaluation. The targeted efficiency improvement suggestions generated by the present invention can significantly improve the energy efficiency and operation efficiency of buildings, which are difficult to achieve by existing technologies.

[0103] Through multi-source data fusion and comprehensively considering various factors, the present invention can be widely applied to various building operation and maintenance scenarios. The targeted efficiency improvement suggestions generated by the present invention can significantly improve the energy efficiency and operation efficiency of buildings and have broad application prospects. In airports, schools, and office buildings, the energy efficiency after optimization is increased by 15% - 25%. Compared with the traditional ANP model, the dynamic weight adjustment reduces the evaluation error by more than 30%.

[0104] The present invention has the following beneficial effects: Determine the operation data set of each electrical device according to the operation and maintenance platform of the building and each electrical device in the building; determine the corresponding low-carbon evaluation model for the electrical device based on the operation data set of each electrical device; determine the corresponding comprehensive weight according to the multiple training of the low-carbon evaluation model of the electrical device and multiple weights in different dimensions; determine the real-time performance score of each electrical device according to the comprehensive weight and the corresponding electrical device, which is compatible with the multiple training of the low-carbon evaluation model of the electrical device and multiple weights in different dimensions, ensures the accuracy of the comprehensive weight, and makes full use of the multiple operations of multiple weights in different dimensions to achieve control from multiple dimensions.

[0105] Determine the overall performance score of the building based on the real-time performance scores of each electrical device and the electricity consumption weights of each electrical device; dynamically optimize the operation strategies of each electrical device according to the overall performance score of the building, present the overall performance effect of the building based on the overall performance score of the building, and dynamically optimize the operation strategies of each electrical device, ensuring the good operation effect of each electrical device and the good effect of the overall performance of the building.

[0106] Please refer to Figure 3 , Figure 3 which is a schematic structural diagram of the overall performance control system of the building in the embodiment of the present invention. The overall performance control system of the building includes: A data set module 21, configured to determine the operation data set of each electrical device according to the operation and maintenance platform of the building and each electrical device in the building; an electrical device low-carbon evaluation module 22, configured to determine the corresponding low-carbon evaluation model for the electrical device based on the operation data set of each electrical device; a comprehensive weight module 23, configured to determine the corresponding comprehensive weight according to the multiple training of the low-carbon evaluation model of the electrical device and multiple weights in different dimensions; a real-time performance scoring module 24 for the electrical device, configured to determine the real-time performance score of each electrical device according to the comprehensive weight and the corresponding electrical device; an overall performance scoring module 25, configured to determine the overall performance score of the building based on the real-time performance scores of each electrical device and the electricity consumption weights of each electrical device; and an optimization module 26, configured to dynamically optimize the operation strategies of each electrical device according to the overall performance score of the building.

[0107] Arbitrarily combine the technical features of the above embodiments. For the sake of brevity of description, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

Claims

1. A method for controlling overall building performance, characterized in that: include: Determine the operation data set of each electrical device according to the building operation and maintenance platform and each electrical device in the building; Determine a corresponding low-carbon evaluation model for the electrical equipment based on the operating data set of each electrical equipment; Determine the corresponding comprehensive weight based on the low-carbon evaluation model of electric equipment and multiple training of weights of multiple different dimensions; Determine the real-time performance score of each electrical device based on the comprehensive weight and the corresponding electrical device; Determine the overall performance score of the building based on the real-time performance score of each electrical device and the power consumption weight of each electrical device; Dynamically optimize the operation strategy of each electrical equipment based on the overall performance score of the building.

2. The building overall performance control method according to claim 1, characterized in that: Determining the operation data set of each electrical device according to the building operation and maintenance platform and each electrical device in the building includes: Collect the building name, and determine the building's operation and maintenance platform based on the building name and building database; Determine each electrical device in the building based on the traversal of the building database; The operation and maintenance platform and each electrical device in the building interact, and the operation data set of each electrical device is determined according to the interaction between the operation and maintenance platform of the building and each electrical device in the building.

3. The building overall performance control method according to claim 1, characterized in that: The step of determining the corresponding low-carbon evaluation model of the electric equipment based on the operation data set of each electric equipment includes: Collecting operation data sets of each electrical device, and determining a plurality of operation data combinations under different operation types based on the detection of the operation data sets of each electrical device; The corresponding low-carbon evaluation model of the electric equipment is determined based on multiple training of multiple operating data combinations, and the low-carbon evaluation results of the corresponding electric equipment are output according to the low-carbon evaluation model of the electric equipment and the corresponding electric equipment, and the operating strategy of the refrigeration host is dynamically adjusted according to the low-carbon evaluation results of each electric equipment.

4. The building overall performance control method according to claim 1, characterized in that: The method of determining the corresponding comprehensive weight according to the low-carbon evaluation model of electric equipment and multiple training of weights of multiple different dimensions includes: In the low-carbon evaluation model of power equipment, an initial judgment matrix is ​​constructed based on the previous operation data of each power equipment, and the network analysis ANP initial weights of each influencing factor are determined according to the initial judgment matrix. ; Calculate the information entropy of each influencing factor, and trigger the correction of the initial weight of each influencing factor according to the information entropy of each influencing factor, so as to correct the weight by outputting the entropy weight method ; Based on the previous operation data of each electrical device, the LSTM model is trained, and the real-time data of the building's operation and maintenance platform is transferred to optimize its prediction accuracy. The mean square error (MSE) is used as the loss function, and the optimizer Adam is used for training to output the prediction results of the LSTM model. Dynamically correct the weights of each influencing factor based on the prediction results of the LSTM model to output the LSTM corrected weights .

5. The building overall performance control method according to claim 4, characterized in that: The method of determining the corresponding comprehensive weight according to the low-carbon evaluation model of the electric equipment and the multiple training of weights of multiple different dimensions also includes: Set the initial weight of ANP , Entropy weight method to correct weights And LSTM modified weights As input, the attention scores of various weights are calculated according to the attention mechanism model; according to the attention scores of various weights, the three weights are weighted and fused to obtain the comprehensive weight .

6. The building overall performance control method according to claim 1, characterized in that: Determining the real-time performance score of each electrical device according to the comprehensive weight and the corresponding electrical device includes: Determine the real-time performance score of each electrical device in the building based on the comprehensive weight and the corresponding electrical equipment ; Real-time performance rating of each electrical device The calculation formula is as follows: 。 7. The building overall performance control method according to any one of claims 1 to 4, characterized in that: The method of determining the overall performance score of the building based on the real-time performance score of each electrical device and the power consumption weight of each electrical device includes: Collect the real-time power consumption data of each power-consuming device, and determine and calculate the power consumption weight of each power-consuming device on the overall performance of the building based on the real-time power consumption data of each power-consuming device. ; Based on the real-time performance score of each electrical equipment and power consumption weight Calculate the overall performance score of the building.

8. The building overall performance control method according to claim 1, characterized in that: The determining of the overall performance score of the building based on the real-time performance score of each electrical device and the power consumption weight of each electrical device also includes: The weight of each electrical device on the overall performance of the building The calculation formula is as follows: ； in: :Electrical equipment Real-time electricity consumption; : The total number of electrical equipment in the building; The overall performance score of a building is calculated using the following formula: ； in: S Maintain the structural attenuation coefficient for the building, S It decays as the building ages.

9. The building overall performance control method according to claim 1, characterized in that: The method of dynamically optimizing the operation strategy of each electrical device according to the overall performance score of the building includes: The overall performance of the building is divided into multiple levels according to the overall performance score of the building; Dynamically optimize the operation strategy of each electrical device based on its real-time performance score and the overall performance of the building.

10. A building overall performance control system, characterized in that: The overall performance control system of the building is applied to the overall performance control method of the building as claimed in any one of claims 1 to 9, and the overall performance control system of the building comprises: A data collection module, used to determine the operation data collection of each electrical device according to the operation and maintenance platform of the building and each electrical device in the building; The low-carbon evaluation module for electric equipment is used to determine the corresponding low-carbon evaluation model for electric equipment based on the operation data set of each electric equipment; A comprehensive weight module is used to determine the corresponding comprehensive weight according to the low-carbon evaluation model of electric equipment and multiple training of weights of multiple different dimensions; The real-time performance scoring module of the electric equipment is used to determine the real-time performance score of each electric equipment according to the comprehensive weight and the corresponding electric equipment; An overall performance scoring module, used to determine the overall performance score of the building based on the real-time performance score of each electrical device and the power consumption weight of each electrical device; The optimization module is used to dynamically optimize the operation strategy of each electrical equipment according to the overall performance score of the building.

Images