A smart park energy efficiency management system and method based on IBMS

Abstract

The present application relates to the field of park management technology, and in particular to an intelligent park energy efficiency management system and method based on IBMS. The method includes: acquiring real-time park data; analyzing the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes; analyzing the energy efficiency changes based on the real-time park data to determine the credibility of each energy efficiency change; and determining an energy efficiency management plan based on the credibility. Optimize energy distribution and reduce energy waste. The credibility assessment model helps to identify the reliability of the prediction results and ensure that the prediction results are not blindly trusted. By analyzing the equipment health indicators and the causes of failures, the credibility of each energy efficiency change can be judged more accurately, thereby avoiding incorrect scheduling due to inaccurate predictions. Based on the credibility of the prediction results, a corresponding energy efficiency management plan is generated, which helps to ensure the accuracy and effectiveness of energy scheduling and avoid energy waste and equipment failures.

Description

Technical Field

[0001] The present application relates to the field of park management technology, and in particular to an IBMS-based smart park energy efficiency management system and method. Background Art

[0002] With the rapid development of smart campuses and intelligent buildings, energy efficiency management within these campuses has become a key technology for improving energy efficiency and achieving sustainable development. Intelligent Building Management Systems (IBMS), a core component of smart campuses, have been widely adopted in various smart buildings, enabling real-time monitoring and regulation of energy consumption within smart campuses, particularly the operation of equipment such as air conditioning, lighting, and heating, ventilation, and air conditioning (HVAC).

[0003] With the diversification of park equipment types and usage requirements, traditional IBMS, faced with complex park environments and constantly changing real-time data, leads to energy waste in park management. Therefore, how to optimize park energy management has become a hot topic in current research and practice. Summary of the Invention

[0004] This application provides an IBMS-based smart park energy efficiency management system and method to solve the above problems.

[0005] In a first aspect, the present application provides a method for energy efficiency management of a smart park based on IBMS, the method comprising:

[0006] Acquire real-time park data; analyze the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes;

[0007] Analyze the energy efficiency changes based on the real-time park data and determine the credibility of each energy efficiency change;

[0008] An energy efficiency management plan is determined based on the credibility.

[0009] Through this solution, real-time campus data collection can provide the real-time operating status of various equipment in the park, such as temperature, humidity, equipment operating status, etc., which helps to preset energy efficiency prediction models and equipment scheduling, so as to make decisions based on the latest environment and usage. Presetting energy efficiency prediction models through historical equipment operating data and deep learning algorithms can help to more accurately predict energy demand and consumption trends under different circumstances in the park, and at the same time help to optimize energy distribution and reduce energy waste. The credibility assessment model helps to identify the reliability of the prediction results and ensure that the prediction results are not blindly trusted. By analyzing equipment health indicators and failure causes, the credibility of each energy efficiency change can be more accurately judged, thereby avoiding incorrect scheduling due to inaccurate predictions. Based on the credibility of the prediction results, a corresponding energy efficiency management plan is generated, which helps to ensure the accuracy and effectiveness of energy scheduling and avoid energy waste and equipment failures.

[0010] Optionally, analyzing the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes includes:

[0011] Analyze the real-time park data to determine the park topography, crowd flow heat map, and park equipment information;

[0012] Divide the park into several sub-areas according to the park equipment information;

[0013] Analyze the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area;

[0014] Based on the preset energy efficiency prediction model, energy efficiency changes are predicted according to the regional characteristics of each sub-region.

[0015] This solution comprehensively identifies the park's geographic layout, pedestrian flow distribution, and equipment operating status by determining the park's topographic map, pedestrian flow heat map, and park equipment information, providing basic data for regional division and energy efficiency prediction. Regional division facilitates personalized management of energy efficiency needs in different regions, improving energy utilization efficiency and reducing unnecessary energy waste. By analyzing the park's topographic map and pedestrian flow heat map, the building structure characteristics and movement trajectory characteristics of each sub-region can be identified, allowing for more accurate prediction of each sub-region's energy efficiency needs. The preset energy efficiency prediction model can predict future energy demand and consumption trends based on the regional characteristics of each sub-region, providing a basis for dynamic scheduling and intelligent decision-making.

[0016] Optionally, determining an energy efficiency management plan based on the credibility includes:

[0017] Comparing the credibility with a preset credibility threshold to obtain a comparison result;

[0018] If the comparison result shows that the credibility is higher than the preset credibility threshold, determining an energy efficiency management plan according to the energy efficiency change;

[0019] If the comparison result shows that the credibility is lower than the preset credibility threshold, determining the regional connection between adjacent sub-areas based on the park equipment information;

[0020] Obtain energy efficiency impact knowledge graph;

[0021] determining energy efficiency impact associations based on the regional connections and the energy efficiency impact knowledge graph;

[0022] Determining the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model;

[0023] According to the abnormal causal contribution, the root cause of the abnormality is determined, and according to the abnormal root cause, the energy efficiency management plan is modified to obtain the best management plan for dealing with the abnormality.

[0024] This solution quickly determines the reliability of prediction results by comparing the confidence level against a pre-set confidence threshold, providing a basis for energy efficiency management decisions. If the confidence level exceeds the preset confidence threshold, an energy efficiency management plan is formulated and implemented based on the predicted energy efficiency changes to optimize energy allocation and improve energy efficiency. If the confidence level falls below the preset confidence threshold, the energy efficiency management plan based on the prediction results is suspended, avoiding potential risks and energy waste. By analyzing campus equipment information in adjacent sub-areas, the mutual influence between adjacent sub-areas is identified, providing clues for anomaly handling. Constructing an energy efficiency impact knowledge graph helps identify energy transfer and influence relationships between devices, providing data support for anomaly detection. Identifying energy efficiency impact correlations facilitates a more comprehensive analysis of anomalies and provides more accurate guidance for anomaly handling. Analyzing the causal contribution of each energy efficiency impact to the prediction deviation helps pinpoint the root cause of the anomaly. Identifying the root cause of the anomaly facilitates revision of the energy efficiency management plan. Based on the root cause of the anomaly, equipment operating parameters are adjusted, energy allocation strategies are optimized, or necessary equipment maintenance is performed, resulting in the optimal management plan for addressing the anomaly and ensuring the accuracy and effectiveness of campus energy efficiency management.

[0025] Optionally, analyzing the energy efficiency changes based on the real-time park data to determine the credibility of each energy efficiency change includes:

[0026] Based on the park equipment information, the real-time park data is analyzed to determine the equipment current waveform, compressor start and stop times, and valve opening of each equipment;

[0027] Analyzing the device current waveform by fast Fourier transform to determine the harmonic distortion rate;

[0028] Obtain historical equipment operation data, analyze the equipment's historical operation time, and determine historical fault information;

[0029] Analyze the historical fault information to determine the cause of the fault;

[0030] Determining, based on the fault cause, a correlation between the harmonic distortion rate and each piece of historical fault information;

[0031] Determining a harmonic distortion rate threshold value according to the association relationship;

[0032] Constructing an equipment health index based on the harmonic distortion rate threshold and the number of compressor starts and stops;

[0033] Based on the equipment health index, the energy efficiency changes are analyzed to determine the credibility of each energy efficiency change.

[0034] This solution identifies each device's current waveform, compressor start / stop times, and valve opening, allowing timely detection of equipment operational issues and providing a basis for energy efficiency management and maintenance. By analyzing the device current waveform, abnormalities in equipment operation can be identified, such as an excessively high harmonic distortion rate (HDR), indicating a fault or malfunction. Analysis of historical operating data helps identify long-term equipment operating trends and potential problems, providing data support for predicting equipment failures. Identifying historical fault information helps identify equipment failure patterns, providing a basis for preventive maintenance and fault prediction. Analyzing fault causes facilitates targeted measures, such as replacing equipment components or adjusting operating parameters, to reduce the occurrence of failures. Determining the correlation between the HDR and each historical fault information helps more accurately predict equipment failures and provide guidance for maintenance. Setting HDR thresholds helps promptly detect abnormalities in equipment operation and provides a standard for maintenance and fault prediction. Equipment health indicators help reflect the overall operating status of equipment, providing a basis for maintenance and operational optimization. Based on equipment health indicators, the credibility of energy efficiency changes can be assessed to support energy efficiency management decisions.

[0035] Optionally, determining the energy efficiency impact association according to the regional connection and the energy efficiency impact knowledge graph includes:

[0036] Analyze the energy efficiency impact knowledge graph and the regional connections to determine the device topology relationship;

[0037] Determine an energy transfer path between any two devices based on the device topology relationship;

[0038] Analyzing the real-time park data to determine the real-time environmental parameters of each sub-area;

[0039] Based on a graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate a dynamic energy efficiency impact weight;

[0040] quantifying the energy efficiency correlation strength between adjacent sub-regions according to the dynamic energy efficiency impact weight;

[0041] An energy efficiency impact correlation is determined according to the energy efficiency correlation strength between the adjacent sub-regions.

[0042] This solution analyzes the energy efficiency impact knowledge graph and regional connections, identifies the physical and logical connections between devices, and facilitates the analysis of energy transfer paths. Identifying the energy transfer path between any two devices helps identify how energy is transferred within the campus, providing a basis for energy efficiency analysis and prediction. By analyzing real-time campus data, real-time environmental parameters such as temperature, humidity, and light in each sub-area are collected to provide real-time data support for energy efficiency prediction. Based on the graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate dynamic energy efficiency impact weights, providing a more accurate decision-making basis for energy efficiency management. Through dynamic energy efficiency impact weights, the strength of the energy efficiency correlation between adjacent sub-areas is quantified, which facilitates energy efficiency coordination and optimization between adjacent sub-areas. Identifying energy efficiency impact correlations helps to more comprehensively identify the mutual influence between adjacent sub-areas, providing a basis for energy efficiency management and equipment scheduling, and achieving more accurate energy allocation and equipment control.

[0043] Optionally, determining the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model includes:

[0044] Analyzing the energy efficiency change based on the preset energy efficiency prediction model to determine the prediction deviation;

[0045] A preset causal forest model is retrieved, and based on the preset causal forest model, the abnormal causal contribution of each energy efficiency impact is determined according to the energy efficiency impact association and the prediction deviation.

[0046] This solution analyzes energy efficiency changes and identifies the energy efficiency performance of equipment, providing a basis for analyzing prediction deviations, thereby improving the accuracy of the preset energy efficiency prediction model, reducing energy demand, and lowering consumption trends. By determining prediction deviations and identifying the accuracy of prediction results, it provides a basis for anomaly detection and correction, reducing energy waste and the risk of equipment failure. Calling the preset causal forest model helps quantify the contribution of each energy efficiency impact to energy efficiency fluctuations and provides a tool for determining abnormal causal contributions. By determining the abnormal causal contribution of each energy efficiency impact, the degree of influence of each energy efficiency impact on prediction deviations is quantified, providing a scientific basis for abnormality handling and energy efficiency management, thereby improving the intelligence level of equipment.

[0047] Optionally, constructing the equipment health index according to the harmonic distortion rate threshold and the number of compressor starts and stops includes:

[0048] Obtain compressor bearing vibration spectrum and refrigerant pressure curve;

[0049] Analyzing the vibration spectrum to determine the level of mechanical wear;

[0050] Analyzing the pressure curve to determine the fluctuation amplitude at the start and stop moments;

[0051] Calculating a sealing performance attenuation coefficient based on the fluctuation amplitude;

[0052] Establish a three-dimensional health assessment space, where the X-axis is the harmonic distortion rate, the Y-axis is the mechanical wear level, and the Z-axis is the sealing performance attenuation coefficient;

[0053] Determining a health benchmark cluster at a current moment according to the three-dimensional health assessment space;

[0054] Calculate the deviation between the current device state and the health reference cluster by Mahalanobis distance;

[0055] A device health indicator is generated based on the deviation.

[0056] This solution acquires detailed equipment operation data by collecting compressor bearing vibration spectra and refrigerant pressure curves in real time. By analyzing the vibration spectrum, the level of mechanical wear is determined, thereby predicting potential equipment failures and maintenance requirements. The fluctuation amplitude during startup and shutdown is determined to assess seal performance, providing a basis for seal maintenance and optimization. The seal performance attenuation coefficient is calculated, helping to predict performance trends and prevent seal failures. By establishing a three-dimensional health assessment space, a more comprehensive assessment of equipment health is achieved, providing a multi-dimensional reference for equipment maintenance and operation optimization. Based on this three-dimensional health assessment space, a health benchmark cluster—the distribution of equipment health under normal conditions—is determined, providing a standard for evaluating equipment status. The Mahalanobis distance is used to calculate the deviation between the current equipment status and the health benchmark cluster, quantifying the degree of abnormality in the equipment status and providing a basis for fault warning and maintenance decisions. Based on the deviation, an equipment health index is generated, reflecting the overall health of the equipment and providing a scientific basis for equipment maintenance and operation optimization.

[0057] Optionally, the modifying the energy efficiency management solution according to the root cause of the abnormality includes:

[0058] Get park event schedules and weather forecast data;

[0059] Determine the scheduled activities based on the park activity schedule and weather forecast data;

[0060] Analyze the temporal and spatial overlap between the root cause of the anomaly and the schedule activities to determine the human influencing factors;

[0061] generating a multi-objective optimization strategy based on the human influencing factors and the weather forecast data;

[0062] Equipment operating parameters and regional linkage rules are adjusted according to the multi-objective optimization strategy.

[0063] This solution uses the campus activity schedule and weather forecast data to better predict and plan energy demand, thereby optimizing energy distribution and equipment scheduling. Based on the campus activity schedule and weather forecast data, upcoming scheduled activities are identified to predict energy demand and equipment usage within different time periods. By analyzing the spatiotemporal overlap between the root causes of anomalies and scheduled activities, human influence factors are identified and targeted corrective measures are taken. Based on human influence factors and weather forecast data, multi-objective optimization strategies are generated, such as adjusting equipment operating parameters and optimizing energy distribution, to improve energy utilization efficiency. Based on the multi-objective optimization strategy, equipment operating parameters and regional linkage rules are adjusted to optimize energy efficiency and improve equipment operating efficiency and stability.

[0064] Optionally, analyzing the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area includes:

[0065] Analyze the topographic map of the park to determine the structural characteristics of the buildings;

[0066] Determine the distribution of ventilation corridors according to the structural characteristics of the building;

[0067] Extract the movement trajectory characteristics of the crowd flow heat map and determine the distribution of people's stay time;

[0068] Calculating the heat load index of each sub-area according to the ventilation corridor distribution and the occupant residence time distribution;

[0069] Determining thermal parameters of the building envelope structure based on the building structure characteristics;

[0070] Correcting the heat load index according to the thermal parameters;

[0071] Based on the modified heat load index, a regional characteristic description vector including a dynamic weight coefficient is generated to obtain the regional characteristics of each sub-region.

[0072] This solution analyzes the park's topographic map and identifies building structural characteristics, facilitating ventilation corridor distribution analysis and heat load calculation. Based on building structural characteristics, the distribution of ventilation corridors within the park is determined, facilitating natural ventilation and air conditioning design. By analyzing the crowd flow heat map and extracting the characteristics of personnel movement trajectories, it facilitates analysis of personnel flow and residence time. Based on the crowd flow heat map, the distribution of personnel residence time within each sub-area within the park is determined, helping to assess the heat load demand of each sub-area. Combining the ventilation corridor distribution and the distribution of personnel residence time, the heat load index of each sub-area is calculated, providing a basis for energy efficiency management. Based on the building structural characteristics, the thermal parameters of the building envelope are determined, providing data support for the correction of the heat load index. Based on the thermal parameters of the building envelope, the heat load index is corrected to improve the accuracy and practicality of the building envelope. Based on the corrected heat load index, a regional characteristic description vector containing a dynamic weight coefficient is generated to reflect the characteristics of each sub-area, providing a basis for personalized energy efficiency management.

[0073] In a second aspect, the present application provides an IBMS-based smart park energy efficiency management system, the system comprising:

[0074] A data analysis module is used to obtain real-time park data; based on a preset energy efficiency prediction model, the real-time park data is analyzed to predict energy efficiency changes;

[0075] A change analysis module, configured to analyze the energy efficiency changes based on the real-time park data and determine the credibility of each energy efficiency change;

[0076] A solution determination module is used to determine an energy efficiency management solution based on the credibility.

[0077] Optionally, the data analysis module analyzes the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes, and is used to:

[0078] Analyze the real-time park data to determine the park topography, crowd flow heat map, and park equipment information;

[0079] Divide the park into several sub-areas according to the park equipment information;

[0080] Analyze the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area;

[0081] Based on the preset energy efficiency prediction model, energy efficiency changes are predicted according to the regional characteristics of each sub-region.

[0082] Optionally, when the solution determination module determines the energy efficiency management solution based on the credibility, it is configured to:

[0083] Comparing the credibility with a preset credibility threshold to obtain a comparison result;

[0084] If the comparison result shows that the credibility is higher than the preset credibility threshold, determining an energy efficiency management plan according to the energy efficiency change;

[0085] If the comparison result shows that the credibility is lower than the preset credibility threshold, determining the regional connection between adjacent sub-areas based on the park equipment information;

[0086] Obtain energy efficiency impact knowledge graph;

[0087] determining energy efficiency impact associations based on the regional connections and the energy efficiency impact knowledge graph;

[0088] Determining the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model;

[0089] According to the abnormal causal contribution, the root cause of the abnormality is determined, and according to the abnormal root cause, the energy efficiency management plan is modified to obtain the best management plan for dealing with the abnormality.

[0090] Optionally, when the change analysis module analyzes the energy efficiency changes based on the real-time park data and determines the credibility of each energy efficiency change, it is configured to:

[0091] Based on the park equipment information, the real-time park data is analyzed to determine the equipment current waveform, compressor start and stop times, and valve opening of each equipment;

[0092] Analyzing the device current waveform by fast Fourier transform to determine the harmonic distortion rate;

[0093] Obtain historical equipment operation data, analyze the equipment's historical operation time, and determine historical fault information;

[0094] Analyze the historical fault information to determine the cause of the fault;

[0095] Determining, based on the fault cause, a correlation between the harmonic distortion rate and each piece of historical fault information;

[0096] Determining a harmonic distortion rate threshold value according to the association relationship;

[0097] Constructing an equipment health index based on the harmonic distortion rate threshold and the number of compressor starts and stops;

[0098] Based on the equipment health index, the energy efficiency changes are analyzed to determine the credibility of each energy efficiency change.

[0099] Optionally, when the solution determination module determines the energy efficiency impact association based on the regional connection and the energy efficiency impact knowledge graph, it is configured to:

[0100] Analyze the energy efficiency impact knowledge graph and the regional connections to determine the device topology relationship;

[0101] Determine an energy transfer path between any two devices based on the device topology relationship;

[0102] Analyzing the real-time park data to determine the real-time environmental parameters of each sub-area;

[0103] Based on a graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate a dynamic energy efficiency impact weight;

[0104] quantifying the energy efficiency correlation strength between adjacent sub-regions according to the dynamic energy efficiency impact weight;

[0105] An energy efficiency impact correlation is determined according to the energy efficiency correlation strength between the adjacent sub-regions.

[0106] Optionally, when the solution determination module determines the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model, it is configured to:

[0107] Analyzing the energy efficiency change based on the preset energy efficiency prediction model to determine the prediction deviation;

[0108] A preset causal forest model is retrieved, and based on the preset causal forest model, the abnormal causal contribution of each energy efficiency impact is determined according to the energy efficiency impact association and the prediction deviation.

[0109] Optionally, when the change analysis module constructs the equipment health index based on the harmonic distortion rate threshold and the number of compressor starts and stops, it is used to:

[0110] Obtain compressor bearing vibration spectrum and refrigerant pressure curve;

[0111] Analyzing the vibration spectrum to determine the level of mechanical wear;

[0112] Analyzing the pressure curve to determine the fluctuation amplitude at the start and stop moments;

[0113] Calculating a sealing performance attenuation coefficient based on the fluctuation amplitude;

[0114] Establish a three-dimensional health assessment space, where the X-axis is the harmonic distortion rate, the Y-axis is the mechanical wear level, and the Z-axis is the sealing performance attenuation coefficient;

[0115] Determining a health benchmark cluster at a current moment according to the three-dimensional health assessment space;

[0116] Calculate the deviation between the current device state and the health reference cluster by Mahalanobis distance;

[0117] A device health indicator is generated based on the deviation.

[0118] Optionally, when the solution determination module amends the energy efficiency management solution based on the root cause of the abnormality, it is configured to:

[0119] Get park event schedules and weather forecast data;

[0120] Determine the scheduled activities based on the park activity schedule and weather forecast data;

[0121] Analyze the temporal and spatial overlap between the root cause of the anomaly and the schedule activities to determine the human influencing factors;

[0122] generating a multi-objective optimization strategy based on the human influencing factors and the weather forecast data;

[0123] Equipment operating parameters and regional linkage rules are adjusted according to the multi-objective optimization strategy.

[0124] Optionally, when the data analysis module analyzes the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area, it is used to:

[0125] Analyze the topographic map of the park to determine the structural characteristics of the buildings;

[0126] Determine the distribution of ventilation corridors according to the structural characteristics of the building;

[0127] Extract the movement trajectory characteristics of the crowd flow heat map and determine the distribution of people's stay time;

[0128] Calculating the heat load index of each sub-area according to the ventilation corridor distribution and the occupant residence time distribution;

[0129] Determining thermal parameters of the building envelope structure based on the building structure characteristics;

[0130] Correcting the heat load index according to the thermal parameters;

[0131] Based on the modified heat load index, a regional characteristic description vector including a dynamic weight coefficient is generated to obtain the regional characteristics of each sub-region. BRIEF DESCRIPTION OF THE DRAWINGS

[0132] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0133] Figure 1 A schematic diagram of an application scenario provided in one embodiment of the present application;

[0134] Figure 2 A flowchart of an IBMS-based smart park energy efficiency management method provided in one embodiment of the present application;

[0135] Figure 3 A schematic diagram of the structure of an IBMS-based smart park energy efficiency management system provided in one embodiment of the present application. DETAILED DESCRIPTION

[0136] To make the purpose, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0137] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0138] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0139] With the diversification of equipment types and usage requirements in industrial parks, traditional forecasting methods often have difficulty accurately capturing complex energy efficiency fluctuations and uncertainties. Therefore, how to optimize the energy management of industrial parks has become a hot topic in current research and practice.

[0140] Based on this, the present application provides an IBMS-based smart campus energy efficiency management system and method. These systems acquire real-time campus data; analyze real-time campus data based on a preset energy efficiency prediction model to predict energy efficiency changes; analyze energy efficiency changes based on real-time campus data to determine the credibility of each energy efficiency change; and determine an energy efficiency management plan based on the credibility. Real-time campus data collection can provide the real-time operating status of various devices within the campus, such as temperature, humidity, and device operating status. This facilitates the pre-setting of energy efficiency prediction models and equipment scheduling, enabling decisions to be made based on the latest environmental and usage data. Pre-setting energy efficiency prediction models based on historical device operating data and deep learning algorithms helps more accurately predict energy demand and consumption trends under different campus conditions, while also helping to optimize energy distribution and reduce energy waste. A credibility assessment model helps identify the reliability of prediction results, ensuring they are not blindly trusted. By analyzing device health indicators and failure causes, the credibility of each energy efficiency change can be more accurately determined, thereby avoiding scheduling errors caused by inaccurate predictions. Based on the credibility of the prediction results, a corresponding energy efficiency management plan is generated, helping to ensure the accuracy and effectiveness of energy scheduling and avoid energy waste and equipment failures.

[0141] Figure 1 This application provides a schematic diagram of an application scenario. The method provided in this application is applied when performing campus energy efficiency management. Specifically, the method provided in this application is applied to any server, which interacts with a number of sensors and collects campus data in real time through the sensors, thereby making decisions based on the latest environment and usage reflected in the campus data. Generating a corresponding energy efficiency management plan helps ensure the accuracy and effectiveness of energy scheduling and avoid energy waste and equipment failure.

[0142] For specific implementation methods, please refer to the following embodiments.

[0143] Figure 2 This is a flowchart of a method for managing energy efficiency of a smart park based on IBMS provided in one embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes:

[0144] S201. Acquire real-time park data; analyze the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes;

[0145] Real-time park data can be collected from the park, such as temperature, humidity, light, Data such as equipment operating status and equipment energy consumption data extracted in real time from sensors and equipment such as concentration.

[0146] The preset energy efficiency prediction model can be a pre-set model based on historical equipment operating data and deep learning algorithms to predict the energy demand and consumption trends of the park over a period of time. It is pre-stored on the server and called when needed.

[0147] Energy efficiency changes can refer to fluctuations in energy usage efficiency within the park.

[0148] Specifically, temperature, humidity, light, Sensors for measuring concentration and equipment operating status collect real-time park data. Leveraging historical equipment operating data and a pre-set energy efficiency prediction model, the system extracts equipment operating status, energy consumption, and other data from real-time park data. This extracted data is then fed into a pre-trained, pre-set energy efficiency prediction model to predict energy efficiency changes.

[0149] S202. Analyze energy efficiency changes based on real-time park data and determine the credibility of each energy efficiency change;

[0150] Credibility can be the degree of confidence in the energy efficiency prediction results.

[0151] Specifically, the system extracts historical equipment operating status data from real-time campus data. The system analyzes historical equipment operating conditions, such as current waveforms and start / stop counts. It uses methods such as Fast Fourier Transform (FFT) to determine harmonic distortion and assess equipment health indicators. Combined with historical equipment operating data, the system analyzes historical operating hours, identifies historical fault information, and determines the causes of these faults. Based on the equipment health indicator assessment results and fault causes, a credibility assessment model is constructed. Real-time campus data is input into the credibility assessment model to calculate the credibility of each energy efficiency change.

[0152] S203. Determine an energy efficiency management plan based on the credibility.

[0153] The energy efficiency management plan can be an energy utilization optimization plan developed for the park based on energy efficiency prediction results and credibility assessment.

[0154] Specifically, a confidence threshold is set to determine whether the prediction results are reliable. The credibility of each energy efficiency change is compared with the preset credibility threshold to determine the reliability of the prediction results. If the credibility is higher than the preset credibility threshold, the prediction results are determined to be reliable, and an energy efficiency management plan is formulated based on the prediction results.

[0155] Through this solution, real-time campus data collection can provide the real-time operating status of various equipment in the park, such as temperature, humidity, equipment operating status, etc., which helps to preset energy efficiency prediction models and equipment scheduling, so as to make decisions based on the latest environment and usage. Presetting energy efficiency prediction models through historical equipment operating data and deep learning algorithms can help to more accurately predict energy demand and consumption trends under different circumstances in the park, and at the same time help to optimize energy distribution and reduce energy waste. The credibility assessment model helps to identify the reliability of the prediction results and ensure that the prediction results are not blindly trusted. By analyzing equipment health indicators and failure causes, the credibility of each energy efficiency change can be more accurately judged, thereby avoiding incorrect scheduling due to inaccurate predictions. Based on the credibility of the prediction results, a corresponding energy efficiency management plan is generated, which helps to ensure the accuracy and effectiveness of energy scheduling and avoid energy waste and equipment failures.

[0156] In some embodiments, real-time park data is analyzed to determine the park topography, crowd heat map, and park equipment information; based on the park equipment information, the park is divided into several sub-areas; the park topography and crowd heat map are analyzed to determine the regional characteristics of each sub-area; based on a preset energy efficiency prediction model, energy efficiency changes are predicted according to the regional characteristics of each sub-area.

[0157] The park topographic map can be a plan view of the park's buildings, roads, green areas and other geographical features.

[0158] A crowd heat map can be a data visualization method that uses the depth of color to represent the crowd density in different areas of the park.

[0159] Park equipment information may include the type, model, location, operating status, energy consumption data, and other information of the equipment within the park.

[0160] Sub-areas can be divided into several small areas based on the park's topographic map and according to data such as park equipment information and crowd flow heat map.

[0161] Regional characteristics can be attributes and features unique to each sub-region.

[0162] Specifically, geographic information technology (GIS) is used to obtain a topographic map of the campus, including building layouts, road distribution, and green areas. Surveillance cameras and real-time campus data are used to analyze personnel flow and generate crowd heat maps, showing the distribution of traffic in different areas. IoT technology is used to collect information about campus equipment, including air conditioning, lighting, HVAC status, and energy consumption. Based on this information and the campus topographic map, a clustering algorithm is used to divide the campus into several sub-areas. Building height, orientation, and number of windows in each sub-area are analyzed based on the campus topographic map. The crowd heat map is used to analyze the movement trajectory characteristics of individuals in each sub-area, such as dwell time and peak activity times. The regional characteristics of each sub-area are determined based on building structural characteristics and movement trajectory characteristics. A pre-set energy efficiency prediction model is trained using historical equipment operation data and regional characteristics. Features such as equipment load, environmental parameters, and human flow activity are extracted from real-time campus data and regional characteristics. These extracted features are input into the trained pre-set energy efficiency prediction model to predict energy efficiency changes.

[0163] In the specific implementation method, a regional classification model can be established through the park topographic map, pedestrian flow heat map and park equipment information, and then this regional classification model can be used to achieve regional division, and then regional correlation analysis can be carried out to facilitate the determination of energy efficiency management plans.

[0164] This solution comprehensively identifies the park's geographic layout, pedestrian flow distribution, and equipment operating status by determining the park's topographic map, pedestrian flow heat map, and park equipment information, providing basic data for regional division and energy efficiency prediction. Regional division facilitates personalized management of energy efficiency needs in different regions, improving energy utilization efficiency and reducing unnecessary energy waste. By analyzing the park's topographic map and pedestrian flow heat map, the building structure characteristics and movement trajectory characteristics of each sub-region can be identified, allowing for more accurate prediction of each sub-region's energy efficiency needs. The preset energy efficiency prediction model can predict future energy demand and consumption trends based on the regional characteristics of each sub-region, providing a basis for dynamic scheduling and intelligent decision-making.

[0165] In some embodiments, the credibility is compared with a preset credibility threshold to obtain a comparison result; if the comparison result shows that the credibility is higher than the preset credibility threshold, the energy efficiency management plan is determined according to the energy efficiency change; if the comparison result shows that the credibility is lower than the preset credibility threshold, the regional connection between adjacent sub-areas is determined according to the park equipment information; the energy efficiency impact knowledge graph is obtained; the energy efficiency impact association is determined according to the regional connection and the energy efficiency impact knowledge graph; the abnormal causal contribution of each energy efficiency impact is determined according to the energy efficiency impact association and the preset energy efficiency prediction model; the abnormal root cause is determined according to the abnormal causal contribution, and the energy efficiency management plan is corrected according to the abnormal root cause to obtain the best management plan for dealing with the abnormality.

[0166] The preset credibility threshold may be a pre-set standard for evaluating the reliability of energy efficiency prediction results, which is pre-stored in the server and called when used.

[0167] Adjacent sub-areas may be two or more sub-areas that are geographically adjacent in the park area division.

[0168] Regional connections can be energy transfer and mutual influence between adjacent sub-regions.

[0169] The energy efficiency impact knowledge graph can be a data structure used to represent the relationship between equipment, energy equipment and environmental factors within a park.

[0170] The energy efficiency impact association may be a specific association of energy transfer and mutual influence between devices in the energy efficiency impact knowledge graph.

[0171] Energy efficiency impact can be the impact of equipment operation data, environmental factors or other factors on energy consumption and energy efficiency performance.

[0172] The abnormal causal contribution degree may be the contribution degree of each factor causing the energy efficiency prediction deviation to the abnormal situation.

[0173] Abnormal root causes can be the root cause of energy efficiency forecast deviations.

[0174] The best management plan can be the best plan for optimizing energy use for the park based on the abnormal root causes and energy efficiency prediction results.

[0175] Specifically, a confidence threshold is set based on analysis of historical equipment operating data and model performance evaluation. A confidence assessment model is used to evaluate each energy efficiency change prediction result and calculate its confidence. The calculated confidence is then compared with the preset confidence threshold to obtain a comparison result. If the confidence exceeds the preset confidence threshold, an energy efficiency management plan is developed based on the predicted energy efficiency change, such as adjusting equipment operating parameters and optimizing energy distribution. If the confidence is lower than the preset confidence threshold, the connectivity between adjacent sub-areas is analyzed based on campus equipment information to determine the regional energy connections between these sub-areas. Natural language processing technology is used to extract knowledge information from equipment manuals, technical documentation, and expert knowledge. A knowledge graph construction tool is used to store the extracted knowledge information in a graph format to construct an energy efficiency impact knowledge graph. Energy efficiency impact associations are identified by combining the regional connections with the energy efficiency impact knowledge graph. The prediction results generated by the preset energy efficiency prediction model are compared with actual energy efficiency data to determine prediction deviations. Based on the energy efficiency impact knowledge graph, the energy efficiency impact that caused the prediction deviation is determined. A causal analysis model is used to calculate the abnormal causal contribution of each energy efficiency impact to the prediction deviation. Based on the anomaly causal contribution calculation results, identify the factor with the greatest impact on the forecast deviation, i.e., the root cause of the anomaly. Based on the identified root cause of the anomaly, revise the existing energy efficiency management plan. Evaluate the revised energy efficiency management plan to determine the optimal management solution for the anomaly.

[0176] This solution quickly determines the reliability of prediction results by comparing the confidence level against a pre-set confidence threshold, providing a basis for energy efficiency management decisions. If the confidence level exceeds the preset confidence threshold, an energy efficiency management plan is formulated and implemented based on the predicted energy efficiency changes to optimize energy allocation and improve energy efficiency. If the confidence level falls below the preset confidence threshold, the energy efficiency management plan based on the prediction results is suspended, avoiding potential risks and energy waste. By analyzing campus equipment information in adjacent sub-areas, the mutual influence between adjacent sub-areas is identified, providing clues for anomaly handling. Constructing an energy efficiency impact knowledge graph helps identify energy transfer and influence relationships between devices, providing data support for anomaly detection. Identifying energy efficiency impact correlations facilitates a more comprehensive analysis of anomalies and provides more accurate guidance for anomaly handling. Analyzing the causal contribution of each energy efficiency impact to the prediction deviation helps pinpoint the root cause of the anomaly. Identifying the root cause of the anomaly facilitates revision of the energy efficiency management plan. Based on the root cause of the anomaly, equipment operating parameters are adjusted, energy allocation strategies are optimized, or necessary equipment maintenance is performed, resulting in the optimal management plan for addressing the anomaly and ensuring the accuracy and effectiveness of campus energy efficiency management.

[0177] In some embodiments, based on the park equipment information, real-time park data is analyzed to determine the equipment current waveform, compressor start-stop times and valve opening of each equipment; the equipment current waveform is analyzed through fast Fourier transform to determine the harmonic distortion rate; the equipment historical operation data is obtained, the equipment historical operation time is analyzed, and historical fault information is determined; the historical fault information is analyzed to determine the cause of the fault; based on the cause of the fault, the correlation between the harmonic distortion rate and each historical fault information is determined; based on the correlation, the harmonic distortion rate threshold is determined; based on the harmonic distortion rate threshold and the compressor start-stop times, an equipment health index is constructed; based on the equipment health index, energy efficiency changes are analyzed to determine the credibility of each energy efficiency change.

[0178] The device current waveform may be a curve showing the change of the current over time during the operation of the device.

[0179] The number of compressor starts and stops may be the number of times the compressor starts and stops within a certain period of time.

[0180] Valve opening can be the degree to which a valve is open to control fluid flow.

[0181] Fast Fourier transform may be a mathematical algorithm for converting a time domain signal into a frequency domain signal.

[0182] Harmonic distortion can be the relative content of harmonic components in the equipment voltage waveform.

[0183] The historical operation data of the equipment can be the operation data of the equipment in the past period of time, such as the startup time, operation time, energy consumption data, fault records, etc. This past period of time can be taken based on experience or specified by people.

[0184] The historical operating time of the equipment can be the cumulative operating time of the equipment in the past period of time. This past period of time can be taken based on experience or determined by opinion.

[0185] Historical fault information may include the fault type, occurrence time, duration, maintenance records, and other information of past equipment faults.

[0186] The cause of the failure can be root causes such as equipment aging, insufficient maintenance, improper operation, external environmental factors, etc. that lead to equipment failure.

[0187] The correlation relationship may be a correlation between the harmonic distortion rate and each historical fault information.

[0188] The harmonic distortion rate threshold may be a preset harmonic distortion rate limit value.

[0189] The device health indicator can be a comprehensive indicator used to evaluate the overall operating status of the device.

[0190] Specifically, based on the park's equipment information, sensors and monitoring equipment are used to collect real-time data on each device's current waveform, compressor start / stop times, and valve opening. Fast Fourier transforms are performed on the collected device current waveforms to analyze their frequency components and determine the harmonic distortion rate (HDR). Historical operating data, such as equipment operating hours and fault records, is extracted from the historical database. By analyzing this historical operating data, historical fault information, including the fault type, occurrence time, and duration, is determined. Based on this historical fault information, fault causes, such as equipment aging, insufficient maintenance, and improper operation, are analyzed. For each type of fault, the harmonic distortion rate (HDR) of the relevant equipment at the time of the fault is analyzed. Statistical analysis methods are used to identify the correlation between the HDR and each piece of historical fault information. The correlation between the HDR and each piece of historical fault information is analyzed. Based on the analysis of the HDR-fault correlation, a HDR threshold is calculated. Statistical analysis methods are used to construct a device health indicator based on the HDR threshold and the compressor start / stop times. A credibility assessment model is constructed using a deep learning algorithm. Equipment health indicators and energy efficiency changes are input into the credibility assessment model to calculate the credibility of each energy efficiency change.

[0191] This solution identifies each device's current waveform, compressor start / stop times, and valve opening, allowing timely detection of equipment operational issues and providing a basis for energy efficiency management and maintenance. By analyzing the device current waveform, abnormalities in equipment operation can be identified, such as an excessively high harmonic distortion rate (HDR), indicating a fault or malfunction. Analysis of historical operating data helps identify long-term equipment operating trends and potential problems, providing data support for predicting equipment failures. Identifying historical fault information helps identify equipment failure patterns, providing a basis for preventive maintenance and fault prediction. Analyzing fault causes facilitates targeted measures, such as replacing equipment components or adjusting operating parameters, to reduce the occurrence of failures. Determining the correlation between the HDR and each historical fault information helps more accurately predict equipment failures and provide guidance for maintenance. Setting HDR thresholds helps promptly detect abnormalities in equipment operation and provides a standard for maintenance and fault prediction. Equipment health indicators help reflect the overall operating status of equipment, providing a basis for maintenance and operational optimization. Based on equipment health indicators, the credibility of energy efficiency changes can be assessed to support energy efficiency management decisions.

[0192] In some embodiments, the energy efficiency impact knowledge graph and regional connections are analyzed to determine the device topology relationship; based on the device topology relationship, the energy transfer path between any two devices is determined; real-time campus data is analyzed to determine the real-time environmental parameters of each sub-area; based on the graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate dynamic energy efficiency impact weights; based on the dynamic energy efficiency impact weights, the energy efficiency correlation strength between adjacent sub-areas is quantified; based on the energy efficiency correlation strength between adjacent sub-areas, the energy efficiency impact correlation is determined.

[0193] The device topology relationship can be the physical and logical connection between any two devices in the campus.

[0194] The energy transfer path can be the path along which energy flows between any two devices in the park.

[0195] Real-time environmental parameters can be environmental parameters such as temperature, humidity, and light intensity in the park that affect equipment operation.

[0196] The graph neural network model can be a deep learning model for processing the coupling relationship between energy transfer paths and real-time environmental parameters.

[0197] The coupling relationship can be the mutual dependence and mutual influence between the energy transfer path and the real-time environmental parameters.

[0198] The dynamic energy efficiency impact weight may be the degree of impact of the device on the overall energy efficiency in a real-time operating state.

[0199] The energy efficiency correlation intensity can be the intensity of the energy efficiency impact between adjacent sub-areas in the park.

[0200] Specifically, the energy efficiency impact knowledge graph and regional connections are used to analyze the connection relationships between devices and determine the device topology relationship. Based on the path search algorithm, the energy transfer path between any two devices in the device topology relationship is determined. Real-time campus data is analyzed to collect real-time environmental parameters such as temperature, humidity, light intensity, and carbon dioxide concentration in each sub-area within the park. A graph neural network model is constructed and trained using historical equipment operation data and real-time equipment operation data, so that the graph neural network model learns the coupling relationship between the energy transfer path and real-time environmental parameters. Based on the trained graph neural network model, the coupling relationship is analyzed to generate dynamic energy efficiency impact weights. Based on the dynamic energy efficiency impact weights, the energy efficiency correlation strength between adjacent sub-areas is quantified using the correlation strength coefficient. Based on the energy efficiency correlation strength between adjacent sub-areas, the energy efficiency impact correlation is determined.

[0201] This solution analyzes the energy efficiency impact knowledge graph and regional connections, identifies the physical and logical connections between devices, and facilitates the analysis of energy transfer paths. Identifying the energy transfer path between any two devices helps identify how energy is transferred within the campus, providing a basis for energy efficiency analysis and prediction. By analyzing real-time campus data, real-time environmental parameters such as temperature, humidity, and light in each sub-area are collected to provide real-time data support for energy efficiency prediction. Based on the graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate dynamic energy efficiency impact weights, providing a more accurate decision-making basis for energy efficiency management. Through dynamic energy efficiency impact weights, the strength of the energy efficiency correlation between adjacent sub-areas is quantified, which facilitates energy efficiency coordination and optimization between adjacent sub-areas. Identifying energy efficiency impact correlations helps to more comprehensively identify the mutual influence between adjacent sub-areas, providing a basis for energy efficiency management and equipment scheduling, and achieving more accurate energy allocation and equipment control.

[0202] In some embodiments, based on a preset energy efficiency prediction model, energy efficiency changes are analyzed to determine prediction deviations; a preset causal forest model is called, and based on the preset causal forest model, the abnormal causal contribution of each energy efficiency impact is determined according to the energy efficiency impact correlation and prediction deviation.

[0203] The prediction deviation can be the difference between the result predicted by the preset energy efficiency prediction model and the actual energy efficiency data.

[0204] The preset causal forest model can be a pre-set machine learning model that quantifies the contribution of each energy efficiency impact to energy efficiency fluctuations. The model is pre-stored in the server and called when used.

[0205] Specifically, a preset energy efficiency prediction model is used to analyze energy efficiency changes in campus equipment. The prediction results of the preset energy efficiency prediction model are compared with actual energy efficiency data to determine the prediction deviation. A preset causal forest model is retrieved and the energy efficiency impact associations and prediction deviations are input into the preset causal forest model. The causal forest model is used to analyze the input energy efficiency impact associations and prediction deviations, and the abnormal causal contribution of each energy efficiency impact to the prediction deviation is determined.

[0206] This solution analyzes energy efficiency changes and identifies the energy efficiency performance of equipment, providing a basis for analyzing prediction deviations, thereby improving the accuracy of the preset energy efficiency prediction model, reducing energy demand, and lowering consumption trends. By determining prediction deviations and identifying the accuracy of prediction results, it provides a basis for anomaly detection and correction, reducing energy waste and the risk of equipment failure. Calling the preset causal forest model helps quantify the contribution of each energy efficiency impact to energy efficiency fluctuations and provides a tool for determining abnormal causal contributions. By determining the abnormal causal contribution of each energy efficiency impact, the degree of influence of each energy efficiency impact on prediction deviations is quantified, providing a scientific basis for abnormality handling and energy efficiency management, thereby improving the intelligence level of equipment.

[0207] In some embodiments, the vibration spectrum of the compressor bearing and the refrigerant pressure curve are obtained; the vibration spectrum is analyzed to determine the mechanical wear level; the pressure curve is analyzed to determine the fluctuation amplitude at the start and stop moments; based on the fluctuation amplitude, the sealing performance attenuation coefficient is calculated; a three-dimensional health assessment space is established, in which the X-axis is the harmonic distortion rate, the Y-axis is the mechanical wear level, and the Z-axis is the sealing performance attenuation coefficient; based on the three-dimensional health assessment space, the health reference cluster at the current moment is determined; the deviation between the current equipment state and the health reference cluster is calculated using the Mahalanobis distance; and based on the deviation, an equipment health index is generated.

[0208] The vibration spectrum of the compressor bearing may be a frequency component distribution diagram converted from a vibration signal generated by the compressor bearing during operation through Fourier transform.

[0209] The refrigerant pressure curve may be a curve showing how the pressure of the refrigerant in the compressor device changes over time.

[0210] Mechanical wear level can be the degree of wear of mechanical parts in the equipment.

[0211] The start and stop moments can be the moments when the equipment starts and stops.

[0212] The fluctuation amplitude can be the difference between the maximum value and the minimum value of the refrigerant pressure curve at the start and stop moments.

[0213] The sealing performance degradation coefficient may be a rate at which the refrigerant sealing performance decays over time.

[0214] The three-dimensional health assessment space may be a three-dimensional coordinate system used to assess the health status of a device.

[0215] The health benchmark cluster may be the health distribution of the equipment in a normal state in the three-dimensional health assessment space.

[0216] Mahalanobis distance can be a method for calculating the distance between the current device status and the health benchmark cluster in the three-dimensional health assessment space.

[0217] The current device status may be the operating status of the device at the current time point.

[0218] The deviation degree can be the degree of difference between the current device status and the healthy baseline cluster.

[0219] Specifically, sensors and monitoring equipment are used to collect real-time data on the compressor bearing vibration spectrum and refrigerant pressure curve. The vibration spectrum is analyzed using a fast Fourier transform to determine the mechanical wear level. The refrigerant pressure curve is analyzed to determine the difference between the maximum and minimum values ​​at the start-up and shutdown moments, i.e., the fluctuation amplitude. A seal performance degradation model is established using statistical methods. Using this model, the seal performance degradation coefficient is calculated based on the fluctuation amplitude data. A three-dimensional coordinate system, the three-dimensional health assessment space, is established, with the harmonic distortion rate (HDR) on the X-axis, the mechanical wear level on the Y-axis, and the seal performance degradation coefficient on the Z-axis. The collected HDR, mechanical wear level, and seal performance degradation coefficient are mapped into the three-dimensional health assessment space to form a point set representing the current equipment status. A cluster analysis algorithm is used to cluster the point set representing the current equipment status and determine the health benchmark cluster at the current moment. The Mahalanobis distance is used to calculate the distance between the current equipment status and the health benchmark cluster. The calculated distance is analyzed to determine the deviation between the current equipment status and the health benchmark cluster. A calculation method for the equipment health index is defined, and the deviation is converted into the equipment health index based on the defined calculation method.

[0220] This solution acquires detailed equipment operation data by collecting compressor bearing vibration spectra and refrigerant pressure curves in real time. By analyzing the vibration spectrum, the level of mechanical wear is determined, thereby predicting potential equipment failures and maintenance requirements. The fluctuation amplitude during startup and shutdown is determined to assess seal performance, providing a basis for seal maintenance and optimization. The seal performance attenuation coefficient is calculated, helping to predict performance trends and prevent seal failures. By establishing a three-dimensional health assessment space, a more comprehensive assessment of equipment health is achieved, providing a multi-dimensional reference for equipment maintenance and operation optimization. Based on this three-dimensional health assessment space, a health benchmark cluster—the distribution of equipment health under normal conditions—is determined, providing a standard for evaluating equipment status. The Mahalanobis distance is used to calculate the deviation between the current equipment status and the health benchmark cluster, quantifying the degree of abnormality in the equipment status and providing a basis for fault warning and maintenance decisions. Based on the deviation, an equipment health index is generated, reflecting the overall health of the equipment and providing a scientific basis for equipment maintenance and operation optimization.

[0221] In some embodiments, a campus activity schedule and weather forecast data are obtained; schedule activities are determined based on the campus activity schedule and weather forecast data; the spatiotemporal overlap between the root cause of the anomaly and the schedule activities is analyzed to determine the human influencing factors; a multi-objective optimization strategy is generated based on the human influencing factors and weather forecast data; and equipment operating parameters and regional linkage rules are adjusted based on the multi-objective optimization strategy.

[0222] The park activity schedule can be a time schedule for various activities in the park.

[0223] Weather forecast data can be the prediction data about future weather conditions provided by the meteorological department.

[0224] Schedule activities can be meetings, exhibitions, festival celebrations and other activities arranged within the park.

[0225] The spatiotemporal overlap may be the degree of overlap between the abnormal root cause and the schedule activity in time and space.

[0226] Human influencing factors can be the impact of the behavior and activities of people in the park on energy efficiency.

[0227] The multi-objective optimization strategy can be the optimal scheduling strategy formulated while considering objectives such as energy efficiency, cost, and user comfort.

[0228] The device operating parameters may be parameters that can be adjusted during the operation of the device, such as the temperature setting of the air conditioner, the brightness of the lighting, etc.

[0229] Regional linkage rules can be linkage control rules for equipment operation between different areas within a park.

[0230] Specifically, obtain the park's activity schedule and weather forecast data from the park management and meteorological departments. Based on the park's activity schedule and weather forecast data, determine the type, time, location, and expected number of attendees for scheduled activities. Analyze the degree of match between the root cause of the anomaly and the time period of the scheduled activity, i.e., the degree of spatiotemporal overlap. Based on the spatiotemporal overlap analysis results, determine the human influencing factors. Build a multi-objective optimization model using an optimization algorithm. Using this multi-objective optimization model, generate a multi-objective optimization strategy based on human influencing factors and weather forecast data. Based on the multi-objective optimization strategy, adjust equipment operating parameters such as air conditioning temperature settings, lighting brightness, and HVAC operating modes, and optimize regional linkage rules.

[0231] This solution uses the campus activity schedule and weather forecast data to better predict and plan energy demand, thereby optimizing energy distribution and equipment scheduling. Based on the campus activity schedule and weather forecast data, upcoming scheduled activities are identified to predict energy demand and equipment usage within different time periods. By analyzing the spatiotemporal overlap between the root causes of anomalies and scheduled activities, human influence factors are identified and targeted corrective measures are taken. Based on human influence factors and weather forecast data, multi-objective optimization strategies are generated, such as adjusting equipment operating parameters and optimizing energy distribution, to improve energy utilization efficiency. Based on the multi-objective optimization strategy, equipment operating parameters and regional linkage rules are adjusted to optimize energy efficiency and improve equipment operating efficiency and stability.

[0232] In some embodiments, the park topographic map is analyzed to determine the building structure characteristics; based on the building structure characteristics, the ventilation corridor distribution is determined; the movement trajectory characteristics of the crowd heat map are extracted to determine the distribution of people's residence time; based on the ventilation corridor distribution and the distribution of people's residence time, the heat load index of each sub-area is calculated; based on the building structure characteristics, the thermal parameters of the building envelope are determined; based on the thermal parameters, the heat load index is corrected; based on the corrected heat load index, a regional characteristic description vector containing a dynamic weight coefficient is generated to obtain the regional characteristics of each sub-area.

[0233] Building structural characteristics can be the physical structure and design features of a building, such as height, layout, orientation, materials, window size and location.

[0234] The ventilation corridor distribution can be a natural ventilation channel designed within the park.

[0235] The movement trajectory characteristics can be the movement paths and patterns of people in the park.

[0236] The distribution of personnel residence time can be the length and frequency of personnel's residence time in each sub-area of ​​the park.

[0237] The heat load index can be a quantitative indicator used to represent the heat load of each sub-area in the park.

[0238] The building envelope can be the building's exterior walls, roof, floors, windows, and doors.

[0239] Thermal parameters can be parameters that describe the thermal performance of building envelopes.

[0240] The dynamic weight coefficient may be a coefficient used to adjust the influence of different factors on the heat load index.

[0241] The regional characteristic description vector may be a multi-dimensional data structure used to describe the characteristics of each sub-region in the park.

[0242] Specifically, geographic information technology was used to analyze the park's topographic map to identify building structural features such as building height, layout, and orientation. Building information models were used to analyze the building's structure. Building ventilation requirements were assessed to determine the need for ventilation corridors. Based on the building's structural characteristics and the need for ventilation corridors, ventilation corridor distribution was designed. Movement trajectory characteristics, such as speed, direction, and frequency, were extracted from the crowd flow heat map. These movement trajectory characteristics were analyzed to determine the dwell time distribution of individuals within each sub-area within the park. A heat load calculation model was established and used to calculate the heat load index for each sub-area based on the distribution of ventilation corridors and dwell time. Based on the building's structural characteristics, thermal parameters such as thermal conductivity, thermal resistance, heat capacity, and solar heat gain coefficient of the building envelope were identified. The heat load calculation model was updated, integrating the collected thermal parameters into the model. The updated heat load calculation model was used to recalculate the heat load index for each sub-area. Based on the results of the heat load calculation model, the heat load index was revised. A multidimensional feature vector was constructed based on the revised heat load index. Data analysis methods were used to determine the dynamic weight coefficient of each feature in the multi-dimensional feature vector. According to the determined dynamic weight coefficient, a regional characteristic description vector including the dynamic weight coefficient is generated. According to the generated regional characteristic description vector, the regional characteristics of each sub-region are obtained.

[0243] This solution analyzes the park's topographic map and identifies building structural characteristics, facilitating ventilation corridor distribution analysis and heat load calculation. Based on building structural characteristics, the distribution of ventilation corridors within the park is determined, facilitating natural ventilation and air conditioning design. By analyzing the crowd flow heat map and extracting the characteristics of personnel movement trajectories, it facilitates analysis of personnel flow and residence time. Based on the crowd flow heat map, the distribution of personnel residence time within each sub-area within the park is determined, helping to assess the heat load demand of each sub-area. Combining the ventilation corridor distribution and the distribution of personnel residence time, the heat load index of each sub-area is calculated, providing a basis for energy efficiency management. Based on the building structural characteristics, the thermal parameters of the building envelope are determined, providing data support for the correction of the heat load index. Based on the thermal parameters of the building envelope, the heat load index is corrected to improve the accuracy and practicality of the building envelope. Based on the corrected heat load index, a regional characteristic description vector containing a dynamic weight coefficient is generated to reflect the characteristics of each sub-area, providing a basis for personalized energy efficiency management.

[0244] Figure 3 A schematic diagram of the structure of a smart park energy efficiency management system based on IBMS is provided in one embodiment of the present application. Figure 3 As shown, the IBMS-based smart park energy efficiency management system 300 of this embodiment includes: a data analysis module 301, a change analysis module 302, and a solution determination module 303.

[0245] The data analysis module 301 is used to obtain real-time park data; analyze the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes;

[0246] A change analysis module 302 is configured to analyze the energy efficiency changes based on the real-time park data and determine the credibility of each energy efficiency change;

[0247] The solution determination module 303 is configured to determine an energy efficiency management solution based on the credibility.

[0248] Optionally, the data analysis module 301 analyzes the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes, and is used to:

[0249] Analyze the real-time park data to determine the park topography, crowd flow heat map, and park equipment information;

[0250] Divide the park into several sub-areas according to the park equipment information;

[0251] Analyze the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area;

[0252] Based on the preset energy efficiency prediction model, energy efficiency changes are predicted according to the regional characteristics of each sub-region.

[0253] Optionally, when determining the energy efficiency management solution based on the credibility, the solution determination module 303 is configured to:

[0254] Comparing the credibility with a preset credibility threshold to obtain a comparison result;

[0255] If the comparison result shows that the credibility is higher than the preset credibility threshold, determining an energy efficiency management plan according to the energy efficiency change;

[0256] If the comparison result shows that the credibility is lower than the preset credibility threshold, determining the regional connection between adjacent sub-areas based on the park equipment information;

[0257] Obtain energy efficiency impact knowledge graph;

[0258] determining energy efficiency impact associations based on the regional connections and the energy efficiency impact knowledge graph;

[0259] Determining the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model;

[0260] According to the abnormal causal contribution, the root cause of the abnormality is determined, and according to the abnormal root cause, the energy efficiency management plan is modified to obtain the best management plan for dealing with the abnormality.

[0261] Optionally, when the change analysis module 302 analyzes the energy efficiency changes based on the real-time park data and determines the credibility of each energy efficiency change, it is configured to:

[0262] Based on the park equipment information, the real-time park data is analyzed to determine the equipment current waveform, compressor start and stop times, and valve opening of each equipment;

[0263] Analyzing the device current waveform by fast Fourier transform to determine the harmonic distortion rate;

[0264] Obtain historical equipment operation data, analyze the equipment's historical operation time, and determine historical fault information;

[0265] Analyze the historical fault information to determine the cause of the fault;

[0266] Determining, based on the fault cause, a correlation between the harmonic distortion rate and each piece of historical fault information;

[0267] Determining a harmonic distortion rate threshold value according to the association relationship;

[0268] Constructing an equipment health index based on the harmonic distortion rate threshold and the number of compressor starts and stops;

[0269] Based on the equipment health index, the energy efficiency changes are analyzed to determine the credibility of each energy efficiency change.

[0270] Optionally, when determining the energy efficiency impact association based on the regional connection and the energy efficiency impact knowledge graph, the solution determination module 303 is configured to:

[0271] Analyze the energy efficiency impact knowledge graph and the regional connections to determine the device topology relationship;

[0272] Determine an energy transfer path between any two devices based on the device topology relationship;

[0273] Analyzing the real-time park data to determine the real-time environmental parameters of each sub-area;

[0274] Based on a graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate a dynamic energy efficiency impact weight;

[0275] quantifying the energy efficiency correlation strength between adjacent sub-regions according to the dynamic energy efficiency impact weight;

[0276] An energy efficiency impact correlation is determined according to the energy efficiency correlation strength between the adjacent sub-regions.

[0277] Optionally, when determining the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model, the solution determination module 303 is configured to:

[0278] Analyzing the energy efficiency change based on the preset energy efficiency prediction model to determine the prediction deviation;

[0279] A preset causal forest model is retrieved, and based on the preset causal forest model, the abnormal causal contribution of each energy efficiency impact is determined according to the energy efficiency impact association and the prediction deviation.

[0280] Optionally, when constructing the equipment health index based on the harmonic distortion rate threshold and the number of compressor starts and stops, the change analysis module 302 is configured to:

[0281] Obtain compressor bearing vibration spectrum and refrigerant pressure curve;

[0282] Analyzing the vibration spectrum to determine the level of mechanical wear;

[0283] Analyzing the pressure curve to determine the fluctuation amplitude at the start and stop moments;

[0284] Calculating a sealing performance attenuation coefficient based on the fluctuation amplitude;

[0285] Establish a three-dimensional health assessment space, where the X-axis is the harmonic distortion rate, the Y-axis is the mechanical wear level, and the Z-axis is the sealing performance attenuation coefficient;

[0286] Determining a health benchmark cluster at a current moment according to the three-dimensional health assessment space;

[0287] Calculate the deviation between the current device state and the health reference cluster by Mahalanobis distance;

[0288] A device health indicator is generated based on the deviation.

[0289] Optionally, when the solution determination module 303 amends the energy efficiency management solution according to the abnormal root cause, it is configured to:

[0290] Get park event schedules and weather forecast data;

[0291] Determine the scheduled activities based on the park activity schedule and weather forecast data;

[0292] Analyze the temporal and spatial overlap between the root cause of the anomaly and the schedule activities to determine the human influencing factors;

[0293] generating a multi-objective optimization strategy based on the human influencing factors and the weather forecast data;

[0294] Equipment operating parameters and regional linkage rules are adjusted according to the multi-objective optimization strategy.

[0295] Optionally, when the data analysis module 301 analyzes the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area, it is used to:

[0296] Analyze the topographic map of the park to determine the structural characteristics of the buildings;

[0297] Determine the distribution of ventilation corridors according to the structural characteristics of the building;

[0298] Extract the movement trajectory characteristics of the crowd flow heat map and determine the distribution of people's stay time;

[0299] Calculating the heat load index of each sub-area according to the ventilation corridor distribution and the occupant residence time distribution;

[0300] Determining thermal parameters of the building envelope structure based on the building structure characteristics;

[0301] Correcting the heat load index according to the thermal parameters;

[0302] Based on the modified heat load index, a regional characteristic description vector including a dynamic weight coefficient is generated to obtain the regional characteristics of each sub-region.

[0303] The system of this embodiment can be used to execute the method of any of the above embodiments. Its implementation principles and technical effects are similar and will not be described in detail here.

Claims

1. A smart park energy efficiency management method based on IBMS, characterized in that: include: Get real-time park data; Analyze the real-time park data based on a preset energy efficiency prediction model to predict energy efficiency changes; Analyze the energy efficiency changes based on the real-time park data and determine the credibility of each energy efficiency change; Determine an energy efficiency management plan based on the credibility; The step of analyzing the real-time park data based on a preset energy efficiency prediction model and predicting energy efficiency changes includes: Analyze the real-time park data to determine the park topography, crowd flow heat map, and park equipment information; Divide the park into several sub-areas according to the park equipment information; Analyze the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area; Based on the preset energy efficiency prediction model, predict energy efficiency changes according to the regional characteristics of each sub-region; Analyzing the energy efficiency changes based on the real-time park data and determining the credibility of each energy efficiency change includes: Based on the park equipment information, the real-time park data is analyzed to determine the equipment current waveform, compressor start and stop times, and valve opening of each equipment; Analyzing the device current waveform by fast Fourier transform to determine the harmonic distortion rate; Obtain historical equipment operation data, analyze the equipment's historical operation time, and determine historical fault information; Analyze the historical fault information to determine the cause of the fault; Determining, based on the fault cause, a correlation between the harmonic distortion rate and each piece of historical fault information; Determining a harmonic distortion rate threshold value according to the association relationship; Constructing an equipment health index based on the harmonic distortion rate threshold and the number of compressor starts and stops; Build a credibility assessment model based on the equipment health indicator evaluation results and fault causes; Input real-time park data into the credibility assessment model to calculate the credibility of each energy efficiency change; Determining an energy efficiency management plan based on the credibility includes: Comparing the credibility with a preset credibility threshold to obtain a comparison result; If the comparison result shows that the credibility is higher than the preset credibility threshold, determining an energy efficiency management plan according to the energy efficiency change; If the comparison result shows that the credibility is lower than the preset credibility threshold, determining the regional connection between adjacent sub-areas based on the park equipment information; Obtain energy efficiency impact knowledge graph; determining energy efficiency impact associations based on the regional connections and the energy efficiency impact knowledge graph; Determining the abnormal causal contribution of each energy efficiency impact based on the energy efficiency impact association and the preset energy efficiency prediction model; According to the abnormal causal contribution, the root cause of the abnormality is determined, and according to the abnormal root cause, the energy efficiency management plan is modified to obtain the best management plan for dealing with the abnormality.

2. The method according to claim 1, characterized in that The determining of energy efficiency impact associations based on the regional connections and the energy efficiency impact knowledge graph includes: Analyze the energy efficiency impact knowledge graph and the regional connections to determine the device topology relationship; Determine an energy transfer path between any two devices based on the device topology relationship; Analyzing the real-time park data to determine the real-time environmental parameters of each sub-area; Based on a graph neural network model, the coupling relationship between the energy transfer path and the real-time environmental parameters is analyzed to generate a dynamic energy efficiency impact weight; quantifying the energy efficiency correlation strength between adjacent sub-regions according to the dynamic energy efficiency impact weight; An energy efficiency impact correlation is determined according to the energy efficiency correlation strength between the adjacent sub-regions.

3. The method according to claim 1, characterized in that The determining of the abnormal causal contribution of each energy efficiency impact according to the energy efficiency impact association and the preset energy efficiency prediction model includes: Analyzing the energy efficiency change based on the preset energy efficiency prediction model to determine the prediction deviation; A preset causal forest model is retrieved, and based on the preset causal forest model, the abnormal causal contribution of each energy efficiency impact is determined according to the energy efficiency impact association and the prediction deviation.

4. The method according to claim 1, wherein The constructing of the equipment health index according to the harmonic distortion rate threshold and the number of compressor starts and stops includes: Obtain compressor bearing vibration spectrum and refrigerant pressure curve; Analyzing the vibration spectrum to determine the level of mechanical wear; Analyzing the pressure curve to determine the fluctuation amplitude at the start and stop moments; Calculating a sealing performance attenuation coefficient based on the fluctuation amplitude; Establish a three-dimensional health assessment space, where the X-axis is the harmonic distortion rate, the Y-axis is the mechanical wear level, and the Z-axis is the sealing performance attenuation coefficient; Determining a health benchmark cluster at a current moment according to the three-dimensional health assessment space; Calculate the deviation between the current device state and the health reference cluster by Mahalanobis distance; A device health indicator is generated based on the deviation.

5. The method according to claim 1, characterized in that The step of modifying the energy efficiency management plan according to the root cause of the abnormality includes: Get park event schedules and weather forecast data; Determine the scheduled activities based on the park activity schedule and weather forecast data; Analyze the temporal and spatial overlap between the root cause of the anomaly and the schedule activities to determine the human influencing factors; generating a multi-objective optimization strategy based on the human influencing factors and the weather forecast data; Equipment operating parameters and regional linkage rules are adjusted according to the multi-objective optimization strategy.

6. The method according to claim 5, characterized in that The analyzing the park topographic map and the pedestrian flow heat map to determine the regional characteristics of each sub-area includes: Analyze the topographic map of the park to determine the structural characteristics of the buildings; Determine the distribution of ventilation corridors according to the structural characteristics of the building; Extract the movement trajectory characteristics of the crowd flow heat map and determine the distribution of people's stay time; Calculating the heat load index of each sub-area according to the ventilation corridor distribution and the occupant residence time distribution; Determining thermal parameters of the building envelope structure based on the building structure characteristics; Correcting the heat load index according to the thermal parameters; Based on the modified heat load index, a regional characteristic description vector including a dynamic weight coefficient is generated to obtain the regional characteristics of each sub-region.

7. A smart park energy efficiency management system based on IBMS, characterized by: The method according to any one of claims 1 to 6 comprises: A data analysis module is used to obtain real-time park data; based on a preset energy efficiency prediction model, the real-time park data is analyzed to predict energy efficiency changes; A change analysis module, configured to analyze the energy efficiency changes based on the real-time park data and determine the credibility of each energy efficiency change; A solution determination module is used to determine an energy efficiency management solution based on the credibility.

Images