Building electromechanical system health monitoring method based on digital twinborn technology

Abstract

The invention provides a building electromechanical system health monitoring method based on a digital twinning technology, and relates to the technical field of electromechanical system monitoring, and the method comprises the steps: obtaining the replacement time and working data of each vulnerable part, obtaining the equivalent life loss rate of each vulnerable part in each task, and forming a life loss digital twinning model; constructing a total life score according to current working data, forming a fatigue coefficient according to system parameters and historical data, obtaining a predicted life loss rate and a predicted wear score according to next operation, constructing a health state score, and judging a health state in combination with a threshold value. The improvement effect on the health condition of the air conditioning system due to replacement of the vulnerable parts is reflected, different service life states of different vulnerable parts are considered during task execution, and the health condition of the air conditioning system is judged more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of electromechanical system monitoring, and specifically to a health monitoring method for building electromechanical systems based on digital twin technology. Background Technique

[0002] The building electromechanical system is an indispensable part of modern buildings, including multiple subsystems such as power supply, water supply, heating, ventilation, and air conditioning (HVAC), and fire protection. The health monitoring methods for different subsystems are also different. Especially for the air conditioning system, it has a complex structure, high working intensity, and the working conditions it faces are ever-changing. During long-term use, some components will experience a reduction in lifespan due to long-term use. For example, the air filter, as the operation of the air conditioning system accumulates, it will affect the normal working effect of the air conditioner due to dust and pollutants, so it needs to be replaced regularly. For example, the fan bearings and belts have their own service life limitations, and when their service life reaches the required level, they need to be replaced. Moreover, different components are replaced at different times, and different new and old components have different impacts on the working ability of the air conditioning system. Therefore, it is necessary to conduct separate analyses on the components replaced at different times in the air conditioner to form a more accurate health monitoring method for the air conditioning system.

[0003] In the prior art, the published document with the publication number CN119397887A discloses an online simulation method and system for wear testing of electromechanical products. It collects historical multi-dimensional monitoring data during the operation of electromechanical products from the historical database; establishes an initial wear prediction model for electromechanical products based on Bayesian networks and convolutional neural networks; trains and tests the initial wear prediction model for electromechanical products through historical multi-dimensional monitoring data to obtain an online wear prediction model for electromechanical products; and real-time collects the multi-dimensional monitoring data of the operation of electromechanical products during the operation of electromechanical products, inputs the multi-dimensional monitoring data of the operation of electromechanical products into the online wear prediction model for electromechanical products, and outputs the wear prediction result of electromechanical products.

[0004] Although the disclosed document realizes the wear monitoring and evaluation of electromechanical systems, it does not consider the changes in health caused by the replacement of vulnerable components in the system, cannot reflect the improvement effect of system maintenance and repair on the system, and does not consider the specific impact of actual operation on the system.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and therefore it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a health monitoring method for building electromechanical systems based on digital twin technology to solve the problems raised in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions: A health monitoring method for a building mechanical and electrical system based on digital twin technology, the specific steps including: Step 1: Obtain the replacement time of the vulnerable components of the air conditioning system, obtain the working data of the air conditioning system and the working data timestamp, the working data including the running power and running duration of each task, compare the replacement time of the vulnerable components with the timestamp of the working data, and obtain the sample set of the tasks experienced by each vulnerable component after each replacement; Step 2: Respectively obtain the unit life loss rate of each replacement of each vulnerable component according to the sample set of the tasks experienced by each vulnerable component after each replacement, and take the unit life loss rate in all replacement processes of each vulnerable component as the equivalent life loss rate of this vulnerable component, summarize the equivalent life loss rates of all vulnerable components, and construct a life loss digital twin model; Step 3: Respectively obtain the time of the last replacement of each vulnerable component through the maintenance log, and the working data experienced since the replacement, input the working data into the life loss digital twin model, respectively obtain the life loss rate of each task of each vulnerable component, and form a life loss sequence of each vulnerable component, combine the life loss sequence of each vulnerable component and the number of vulnerable components with high loss to construct an overall life score; Step 4: According to the parameters and historical data of the air conditioning system, screen the tasks with running power exceeding the rated power and their running duration, and the tasks with running power lower than 70% of the rated power, and obtain the average power of these tasks, and construct a system fatigue coefficient; Step 5: Obtain the next running power and next running duration according to the next temperature control task to be executed by the air conditioning system, input them into the life loss digital twin model, respectively obtain the next life loss rate of each vulnerable component, and obtain the corrected next life loss rate through correction by the system fatigue coefficient, add the corrected next life loss rate to the life loss sequence of each vulnerable part to form a predicted life loss sequence, obtain the predicted wear score again according to the overall life score, generate a health score, and judge the state of the air conditioning system according to the relationship between the health score and threshold Ⅰ and threshold Ⅱ.

[0008] Further, obtain the replacement time of the vulnerable components of the air conditioning system, the vulnerable components including air filters, fan bearings, fan belts, compressor valves and condenser pipes, the replacement time is obtained through the maintenance log, obtain the working data of the air conditioning system and the working data timestamp, the working data and timestamp are obtained through the work log, the working data including the running power and running duration of each task, the running power being the running power of the entire air conditioning system, map the working data to the replacement time, and the logic of the mapping is: Compare the replacement time with the timestamp of the working data to obtain the total number of tasks experienced by each vulnerable component after each replacement, as well as the operating power and operating duration of each task. The operating power of each vulnerable component is the operating power of the entire system during this task, and the operating duration of each vulnerable component is the total operating duration of the entire system during this task; Number each vulnerable component, and respectively obtain the sample set of tasks experienced by each vulnerable component after each replacement. The formula is as follows:

[0009] Among them, represents the sample set after the th replacement of the th vulnerable component, represents the operating power of the th task, represents the operating duration of the th task, is the retrieval variable for the vulnerable component number, , , is the retrieval variable for the task number, , , is the total number of tasks after the th replacement of this vulnerable component, is the retrieval variable for the replacement number, , , is the th total replacement number of the

[0010] Furthermore, according to the sample set of each vulnerable component after each replacement, respectively obtain the equivalent life depreciation rate of each vulnerable component. The logic is as follows: Obtain the rated life of each vulnerable component for each replacement. The rated life is the working duration of this vulnerable component under the rated power, obtained from the manufacturer of the replaced vulnerable component. Respectively obtain the unit life depreciation rate of each vulnerable component for each replacement. The unit life depreciation rate of each replacement represents the life depreciation rate of this vulnerable component in terms of unit working efficiency and unit working duration from installation to being replaced. The formula is as follows:

[0011] Among them, represents the unit life depreciation rate after the th replacement of the th vulnerable component, represents the th vulnerable component and the The operating power of the th task after the first replacement, where represents the operating duration of the th task after the first replacement of the ith vulnerable component; , , represents the total number of tasks after the ith replacement of this vulnerable component; , , is the total number of replacements of the ith vulnerable component; , ; Take the unit life loss rate in all replacement processes of each vulnerable component as the equivalent life loss rate of this vulnerable component. The formula is as follows:

[0012] where is the equivalent life loss rate of the ith vulnerable component, represents the equivalent life loss rate after the ith replacement of the ith vulnerable component; , is the replacement times retrieval variable, , is the total number of replacements of the ith vulnerable component;

[0013] Furthermore, obtain the time of the last replacement of each vulnerable component through the maintenance log, and obtain the work data experienced by this vulnerable component since the replacement through the work log. Input the work data into the life loss digital twin model to obtain the life loss rate of each task of each vulnerable component respectively, and form a life loss sequence. The formula is as follows:

[0014] Among them, represents the life loss sequence of the th vulnerable component, is the retrieval variable for the vulnerable component number, , , represents the life loss rate of the th mission, is the retrieval variable for the mission times, , , represents the number of missions experienced.

[0015] Furthermore, based on the life loss sequences of all vulnerable components, an overall life score is constructed, and the formula is as follows:

[0016] Among them, is the overall life score, represents the life loss rate of the th vulnerable component in the th mission, is the retrieval variable for the vulnerable component number, , , is the retrieval variable for the mission times, , , represents the number of missions experienced, represents the high loss index of the th vulnerable component, is a coefficient; Obtain the high loss determination threshold, and the logic for obtaining the high loss index is as follows:

[0017] Among them, represents the high loss index of the th vulnerable component, is the high loss determination threshold, is the retrieval variable for the mission times, , , represents the number of operating times experienced, is the retrieval variable for the vulnerable component number, , .

[0018] Further, obtain the air-conditioning system parameters. The system parameters of the air-conditioning system include the designed life, the average number of replacements of vulnerable components, and the rated power of this air-conditioning system. Obtain the historical data of this air-conditioning system. The historical data includes the operating duration and the total number of replacements. The operating duration is the sum of the operating durations of all tasks recorded in the work log. The total number of replacements is the sum of the number of times of replacing vulnerable components of this air-conditioning system. The number of times of replacing vulnerable components is obtained through the maintenance log. Screen the tasks with operating power exceeding the rated power and their operating durations, and number them. Screen the tasks with operating power lower than 70% of the rated power and obtain the average power of these tasks. Construct the system fatigue coefficient, and the basis formula is as follows:

[0019] Wherein, is the fatigue coefficient, is the operating duration, is the designed life, is the total number of replacements, is the average number of replacements of vulnerable components, represents the th operating power of the task with power exceeding the rated power, is the rated power, is the th operating duration of the task with power exceeding the rated power, is the retrieval variable of the task with power exceeding the rated power, , , is the number of tasks with power exceeding the rated power, is the average power, is the coefficient.

[0020] Further, obtain the next temperature control task of the air-conditioning system, and obtain the next required operating power and the next operating duration of the air-conditioning system according to the temperature control task. The operating power and the operating duration are manually set according to the performance of the air-conditioning system. Input the next operating efficiency and the next operating duration into the life loss digital twin model, obtain the next life loss rate of each vulnerable component respectively, and correct the next life loss rate to obtain the corrected next life loss rate. The basis formula is as follows:

[0021] Wherein, is the corrected next life loss rate, is the next life loss rate, is the fatigue coefficient; Add the corrected next life loss rate into the life loss sequence to form the predicted life loss sequence. The basis formula is as follows:

[0022] Among them, represents the predicted life loss sequence of the th vulnerable component, represents the life loss sequence of the th vulnerable component, is the retrieval variable of the vulnerable component number, , , is the corrected next life loss rate.

[0023] Furthermore, according to the predicted life loss sequence, the overall life score is formed again according to step 3, and the overall life score formed this time is calibrated as the predicted wear score, and the health score is generated. The formula is as follows:

[0024] Among them, is the health score, is the overall life score, is the predicted wear score.

[0025] Furthermore, set threshold I and threshold II, and threshold I is less than threshold II. Analyze according to the health score. The logic is as follows: When the health score is greater than or equal to threshold II, it is determined that the system is healthy, indicating that the next task can be satisfied; When the health score is less than threshold II and greater than threshold I, it is determined that the health level of the system is low, and an early warning is given, indicating that the next task can be satisfied but the vulnerable components need to be replaced after the task ends; When the health score is less than or equal to threshold I, it is determined that there are potential hazards in the system, and an alarm is given, indicating that it is not enough to complete the next task.

[0026] Compared with the prior art, the beneficial effects of the present invention are: By obtaining the replacement time and working data of the vulnerable components in the air conditioning system, the present invention forms the equivalent life loss rate of each vulnerable component for different task amounts, establishes a life loss digital twin model, constructs an overall life score, and judges the health status of the system according to the relationship between the current overall life score and the predicted wear score, reflecting the improvement effect on the health status of the air conditioning system due to the replacement of vulnerable components, taking into account the different life states of different vulnerable components during task execution, and more accurately judging the health status of the air conditioning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0028] To make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0029] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meaning understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not represent any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0030] Embodiment:

[0031] Please refer to Figure 1 , the present invention provides a technical solution: Step 1: Obtain the replacement time of the vulnerable parts of the air conditioning system, obtain the working data of the air conditioning system and the timestamp of the working data. The working data includes the running power and running duration of each task. Compare the replacement time of the vulnerable parts with the timestamp of the working data to obtain a sample set of the tasks experienced after each replacement of each vulnerable part; The said Step 1 includes the following contents: Obtain the replacement time of the vulnerable parts of the air conditioning system. The vulnerable parts include air filters, fan bearings, fan belts, compressor valves and condenser pipes. The replacement time is obtained through maintenance logs. Obtain the working data of the air conditioning system and the timestamp of the working data. The working data and timestamp are obtained through work logs. The working data includes the running power and running duration of each task. The running power is the running power of the entire air conditioning system. Map the working data to the replacement time. The logic of the mapping is: Compare the replacement time with the timestamp of the working data to obtain the total number of tasks experienced after each replacement of each vulnerable part, as well as the running power and running duration of each task. The running power of each vulnerable part is the running power of the entire system during this task, and the running duration of each vulnerable part is the total running duration of the entire system during this task; Air filters, fan bearings, fan belts, compressor valves, and condenser tubes are consumable parts in an air-conditioning system. The air filter will significantly reduce the performance of the air-conditioning system due to dust blockage. Fan bearings and fan belts will cause abnormal operation of the air-conditioning system due to fatigue damage, and may even cause the air-conditioning system to stop working. Compressor valves will cause abnormal operation of the compressor due to wear, resulting in a significant decline in the refrigerant compression effect, thus affecting the performance of the air-conditioning system. Condenser tubes are prone to bacterial growth and blockage due to long-term use, thus affecting the operating state of the air-conditioning system. Therefore, they need to be replaced regularly.

[0032] Number each vulnerable component, and respectively obtain the sample set of the tasks experienced after each replacement of each vulnerable component. The formula is as follows:

[0033] Where, represents the sample set after the th replacement of the th vulnerable component, represents the operating power of the th task, represents the operating duration of the th task, is the retrieval variable for the vulnerable component number, , , are the retrieval variables for the task times, , , is the total number of tasks after the th replacement of this vulnerable component, is the retrieval variable for the replacement times, , , is the th total replacement times of the vulnerable component.

[0034] It is possible to extract the replacement time of vulnerable parts (air filters, fan bearings, fan belts, compressor valves, condenser tubes) from the maintenance log, and extract the timestamp data of the operating power and operating time of each task from the work log, and then establish the task sample set within the replacement cycle of each component through time matching. Through strict time mapping, the precise split between the actual replacement cycle of the parts and the equipment operation data is achieved, ensuring that the subsequent life loss calculation only considers the data during the actual service period, avoiding the interference of data mixing on the result evaluation; this data division method provides a clear interval definition for the subsequent establishment of the digital twin model, so that the state evolution of each component in each replacement cycle is independently and finely described; it not only lays the foundation for the quantification of the state of a single vulnerable component, but also establishes the data origin for the health status assessment of the entire system, thus forming a rigorous and hierarchical closed-loop connection in data transmission and calculation logic, providing a high degree of reliability and consistency for subsequent steps.

[0035] Step 2: Obtain the unit life loss rate of each vulnerable part each time it is replaced based on the sample set of tasks experienced by each vulnerable part after each replacement, and take the unit life loss rate of each vulnerable part in all replacement processes as the equivalent life loss rate of the vulnerable part. Sum up the equivalent life loss rates of all vulnerable parts and build a life loss digital twin model. The step 2 includes the following contents: According to the sample set after each replacement of each vulnerable part, the equivalent life depreciation rate of each vulnerable part is obtained respectively. The logic is as follows: The rated life of each vulnerable part is obtained each time it is replaced. The rated life is the working time of the vulnerable part at the rated power. It is obtained from the manufacturer of the replaced vulnerable part. The unit life depreciation rate of each vulnerable part is obtained each time it is replaced. The unit life depreciation rate of each replacement represents the life depreciation rate of the unit work efficiency and unit working time of the vulnerable part from installation to replacement. The formula for the unit life depreciation rate is as follows:

[0036] in, Indicates Consumable parts Unit life loss rate after replacement, Indicates Consumable parts After the first replacement The operating power of the task, Indicates Consumable parts After the first replacement The running time of the task, Indicates the rated life of the th vulnerable component part, is the retrieval variable for the number of tasks, , , is the total number of tasks after the th replacement of this vulnerable component part, is the retrieval variable for the number of replacements, , , is the total number of replacements of the th vulnerable component part, is the retrieval variable for the serial number of the vulnerable component part, , ; Indicates the unit life depreciation rate of the th vulnerable component part during its th replacement cycle, which reflects the degree to which the component consumes its own life due to bearing the actual load in the actual work tasks from installation to replacement. Counts the cumulative "workload" experienced by the component in all tasks during this replacement cycle, where indicates the operating power of the th task (reflecting the actual load and energy consumption in the working environment), while indicates the operating duration of this task (reflecting the length of the load acting time), and the product of the two can be regarded as a quantitative index of the stress or wear caused by the task to the component; represents the designed life duration of the th component part under the rated power condition specified by the manufacturer, that is, the duration during which the component can work normally under the standard working condition. This formula measures the degree of life depreciation of the component by comparing the total accumulated load in actual work with the theoretical standard life. When the operating power or the operating duration increases for each task, the accumulated workload also rises, resulting in a decrease in the unit life depreciation rate; conversely, if the component mainly works under low power and short duration conditions and still leads to component replacement, its fatigue loss will be relatively large, making relatively high. serves to normalize the actual working state to the designed life. Therefore, when the rated life is relatively long, the accumulated workload is relatively high and the unit life depreciation rate may be relatively low; conversely, it is higher. It reflects the consumption effect of the accumulated load under the actual working conditions on the component life, and through the multiplication and normalization relationship between variables, quantifies the complex operating conditions into an index reflecting the life consumption efficiency, providing an intuitive basis for maintenance and prediction.

[0037] Take the unit life loss rate of each vulnerable component during all replacement processes as the equivalent life loss rate of this vulnerable component. The formula is as follows:

[0038] Where, is the equivalent life loss rate of the th vulnerable component, represents the equivalent life loss rate after the th replacement of the th vulnerable component, is the retrieval variable for the vulnerable component number, , , are the retrieval variables for the replacement times, , , is the total number of replacements of the th vulnerable component; For a component that has been replaced several times, the mission experiences of the components replaced each time may be different, which may lead to deviations in the calculation of the unit life loss rate. Take the average of the unit life loss rates of the components replaced each time as the equivalent life loss rate to reduce the deviations caused by possible quality problems of individual components and facilitate obtaining more accurate data.

[0039] Summarize the equivalent life loss rates of each vulnerable component to construct a life loss digital twin model; By comparing the real mission data with the rated life calibrated by the manufacturer, the accurate docking of "actual consumption and theoretical indicators" is achieved, which can truly reflect the wear intensity of each component under different usage conditions; averaging the data of each replacement cycle not only effectively smooths the accidental effects caused by operation fluctuations, but also improves the consistency and reliability of the data, providing a stable and accurate historical basis for predicting trends; summarizing and digitally twin modeling the life loss rates of each component constructs a unified data platform for the next-step real-time prediction and health scoring, enabling the data transfer of the entire system to achieve seamless docking from local to whole and from history to future, and ensuring the continuity and scientificity of the internal data logic between upstream and downstream steps.

[0040] Step 3: Obtain the time of the last replacement of each vulnerable component and the work data experienced since the replacement through the maintenance log, input the work data into the life loss digital twin model, obtain the life loss rate of each mission of each vulnerable component respectively, and form a life loss sequence of each vulnerable component. Combine the life loss sequences of each vulnerable component and the number of vulnerable components with high loss to construct an overall life score; The said Step 3 includes the following contents: Step 301: Obtain the time when each vulnerable component was last replaced through the maintenance log, and obtain the work data experienced by the vulnerable component since the replacement through the work log. Input the work data into the life loss digital twin model to obtain the life loss rate of each task for each vulnerable component, and form a life loss sequence. The basis formula is as follows:

[0041] Among them, represents the life loss sequence of the th vulnerable component, is the retrieval variable for the vulnerable component number, , , represents the life loss rate of the th task, is the retrieval variable for the task number, , , represents the number of tasks experienced.

[0042] By collecting the task data after the last replacement and inputting the data into the digital twin model, the life loss rate corresponding to each task is obtained, thereby constructing the life loss sequence of each component. This recording method can continuously track the wear evolution process of components, capturing both the changes in instantaneous states and providing a data basis for long-term health trends; the continuous recording provides dynamic monitoring indicators for subsequent life prediction and maintenance decisions.

[0043] Step 302: Construct an overall life score based on the life loss sequences of all vulnerable components. The basis formula is as follows:

[0044] Among them, is the overall life score, represents the life loss rate of the th vulnerable component for the th task, is the retrieval variable for the vulnerable component number, , , is the retrieval variable for the task number, , , represents the number of tasks experienced, represents the high loss index of the th vulnerable component, is the coefficient; Represents the overall life score, quantifying the cumulative wear of all vulnerable components in the entire air conditioning system. The continuous multiplication indicates the product aggregation of all 5 vulnerable components, reflecting the comprehensive impact of the state of each component during system operation on the overall health; and calculates the cumulative life loss rate of the th vulnerable component during the th operation task, where each is the wear impact suffered by the component during a certain task, truly reflecting the load effect during task operation. The parameter plays a role in adjusting the contribution degree of the cumulative damage of a single component, and its magnitude determines the overall impact of the sum of component wear on S. When the sum becomes larger, it indicates that the component has withstood higher loads and wear in the actual working conditions, thus increasing the value of the corresponding part of the component in ; otherwise, it decreases. Introduces the high wear index for each vulnerable component. This index takes 1 when the cumulative wear reaches the preset threshold, and 0 otherwise, truly reflecting whether there is a local severe wear phenomenon in actual work. By taking the exponent after averaging these indexes, the impact of high wear phenomena of several components on the overall score is amplified, enabling potential problems in the system health status to be quickly reflected. reflects the superposition effect of cumulative wear (and local high wear risk), and, as the actual cumulative damage of each component increases, it also increases. At the same time, when the proportion of the high wear index becoming 1 rises, the exponential part will cause to show a non-linear upward trend. Therefore, being higher means that the overall system bears more fatigue and wear, indicating that its health status may be worse, thus providing a reliable quantitative basis for maintenance and early warning decisions.

[0045] Obtain the high wear determination threshold, and the logic for obtaining the high wear index is as follows:

[0046] Among them, represents the high wear index of the th vulnerable component, is the high wear determination threshold, is the task number retrieval variable, , , represents the number of operations experienced, is the vulnerable component number retrieval variable, , .

[0047] Indicates the high wear index of the th vulnerable component, reflecting whether the cumulative life wear rate of the component in the missions it has experienced exceeds a preset high wear determination threshold , so as to judge the state of being at a higher wear risk. Indicates the th component's life wear rate in the th mission, which directly reflects the load impact and wear intensity borne by the component in the actual operating environment. The cumulative sum then reflects the cumulative load loss over the entire service life; the threshold is derived from engineering design and historical data and is used to define the critical point between normal wear and abnormal wear under standard working conditions. When the cumulative wear reaches or exceeds this critical point, the index takes the value of 1, meaning that the component has a higher life wear risk and sufficient attention needs to be paid; conversely, when the cumulative wear is below this threshold, it takes 0, indicating that the component is still in a safe or normal wear range. Converts the continuously generated wear data under complex actual working conditions into an easily recognizable and applicable signal through a simple binary determination method, thus playing a role in amplifying the impact of the high wear state in the subsequent overall life score and health prediction, helping maintenance personnel identify potential problems in a timely manner and make decisions.

[0048] Using the life wear sequences and high wear indices of all components, an overall life score is generated through comprehensive calculation. The high wear index introduced here further amplifies the impact of key components on the overall health of the system under abnormal wear conditions, so that the overall score can more sensitively reflect the threat of local fault evolution to the global health level. Steps 301 and 302 form a logical progressive process from point to line and from local to global, not only quantifying the operating state of each vulnerable component, but also enabling multi-point data to be synchronously compared within the digital twin framework, constituting a data system that links the past and the future and providing a quantitative standard for dynamic evolution for system health prediction.

[0049] Step 4: According to the parameters of the air conditioning system and historical data, screen the missions with operating power exceeding the rated power, their operating durations, and the missions with operating power lower than 70% of the rated power, and obtain the average power of these missions to construct a system fatigue coefficient; The said step 4 includes the following contents: Obtain the system parameters of the air-conditioning system. The system parameters of the air-conditioning system include the designed life, the average number of replacements of vulnerable parts, and the rated power of this air-conditioning system. Obtain the historical data of this air-conditioning system. The historical data includes the operation duration and the total number of replacements. The operation duration is the total operation duration of all tasks recorded in the work log. The total number of replacements is the sum of the number of times of replacing vulnerable parts of this air-conditioning system. The number of times of replacing vulnerable parts is obtained through the maintenance log. Screen the tasks with operating power exceeding the rated power and their operation durations, and number them. Screen the tasks with operating power lower than 70% of the rated power and obtain the average power of these tasks. Construct the system fatigue coefficient, and the basis formula is as follows:

[0050] Among them, is the fatigue coefficient, is the operation duration, is the designed life, is the total number of replacements, is the average number of replacements of vulnerable parts, represents the th operating power of the task with power exceeding the rated power, is the rated power, is the th operation duration of the task with power exceeding the rated power, is the retrieval variable of the task with power exceeding the rated power, , , is the number of tasks with power exceeding the rated power, is the average power, is the coefficient.

[0051] represents the system fatigue coefficient, which comprehensively reflects the additional fatigue and wear caused by factors such as working hours, component replacement frequency, and tasks with power exceeding the rated and low efficiency during the actual operation of the air-conditioning system. represents the actual operation duration of the system, while is the designed life duration. When increases, it means that the cumulative load of the system under normal working conditions also increases. Therefore, decrease will cause to show an upward trend; is the total number of actually replaced components, is the average number of replacements, which is used to measure the impact of replacement frequency on system fatigue. When the number of replacements increases, the value of will increase, thereby increasing the system fatigue coefficient , thus reflecting the consumption of system life caused by frequent component replacement. Accumulated the additional fatigue caused by all tasks exceeding the rated power to the system, where is the operating power of the th task exceeding the rated power, is the operating duration of this task, is the exponential factor for adjusting the effect of overloaded tasks, reflecting the boosting effect on system fatigue when there are more overloaded operating tasks and longer operating durations; while represents the total duration corresponding to tasks with an average operating power lower than 70% of the rated power, and its ratio is used as a subtraction term, meaning that working in a low-power state can relieve system fatigue to a certain extent. Achieved the quantitative synthesis of the impact of different operating conditions on system wear, reflected the combined effect of actual operating duration, component replacement frequency, and the deviation of the working state from the designed operating conditions on system fatigue, and indicated that the equipment may face higher maintenance risks in future operations.

[0052] Linked the local wear conditions of individual components with the overall system operating environment and working mode, systematized the dynamic operating condition parameters affecting system durability and fatigue properties, and formed a conversion channel from single-piece indicators to overall effects; by introducing a fatigue coefficient to correct the additional wear caused by additional operating conditions, it made up for the possible deviation in predicting only relying on the local life loss rate, making subsequent predictions more in line with actual operating conditions; the in-depth investigation of historical data provided a correction factor for comparing future task information with historical trends in the next step, thus playing a role in connecting historical operating states and future task requirements in the entire maintenance decision-making process, and enhancing the adaptability and accuracy of the prediction model.

[0053] Step 5: Obtain the next operating power and next operating duration according to the next temperature control task to be executed by the air conditioning system, input them into the life loss digital twin model, obtain the next life loss rate of each vulnerable component respectively, and obtain the corrected next life loss rate through system fatigue coefficient correction. Add the corrected next life loss rate to the life loss sequence of each vulnerable part to form a predicted life loss sequence. Again, obtain the predicted wear score according to the overall life score, generate a health score, and judge the state of the air conditioning system according to the relationship between the health score and threshold Ⅰ and threshold Ⅱ.

[0054] The said step 5 includes the following contents: Step 501: Obtain the next temperature control task of the air conditioning system, and based on the temperature control task, obtain the next required operating power and operating duration of the air conditioning system. The operating power and operating duration are manually set according to the performance of the air conditioning system. Input the next operating efficiency and operating duration into the life loss digital twin model, obtain the next life loss rate of each vulnerable component respectively, and correct the next life loss rate to obtain the corrected next life loss rate. The basis formula is as follows:

[0055] Wherein, is the corrected next life loss rate, is the next life loss rate, is the fatigue coefficient; The next life loss rate is corrected by the system fatigue coefficient. Since the air conditioning system has been used for a long time, the consumption of its own life means the aging of other components, which affects the normal calculation of the next life loss rate of vulnerable components. Therefore, it is necessary to correct the next life loss rate.

[0056] Add the corrected next life loss rate into the life loss sequence to form a predicted life loss sequence. The basis formula is as follows:

[0057] Wherein, represents the predicted life loss sequence of the th vulnerable component, represents the life loss sequence of the th vulnerable component, is the retrieval variable of the vulnerable component number, , , is the corrected next life loss rate.

[0058] Obtain the expected operating power and duration from the temperature control task that the air conditioning system is about to execute, input these estimated data into the digital twin model, calculate the next life loss rate, and then correct it in combination with the fatigue coefficient, so that future work tasks can be included in the system wear prediction in advance, realizing the key transformation from historical data backtracking to future state preview; the correction of the fatigue coefficient ensures that the impact when the task parameters deviate from the design value can be quantitatively considered, enhancing the realistic matching degree of the prediction result; Step 502: According to the predicted life loss sequence, form the overall life score again according to Step 3, and calibrate the overall life score formed this time as the predicted wear score to generate a health score. The basis formula is as follows:

[0059] Wherein, is the health score, is the overall life score, is the predicted wear score.

[0060] represents the health score, which is used to reflect the quantitative evaluation of the overall health status of the air-conditioning system after calibration of historical wear and predicted wear. is the overall life score, representing the cumulative depreciation situation of the system during historical operation. The higher the value, the more serious the cumulative wear of the system and the relatively lower the remaining life; while is the predicted wear score, reflecting the possible wear trend of the system under the influence of future tasks. First, through obtain a remaining health capacity based on the historical state. When is smaller, is larger, indicating that the historical wear of the system is lighter and the health status is better; subsequently, through the exponential function take into account the increment of predicted wear relative to the current depreciation. If is much higher than (that is, the future wear increases significantly), then tends to be larger, and the exponential part drops sharply, resulting in the overall health score decreasing, thus indicating that there may be problems with the system health status. On the contrary, if is close to or basically the same as , then the exponential term is close to , making remain near , indicating that the future wear prediction of the system is more consistent with the historical state and the health risk is lower. By combining the current health state with the future wear prediction and amplifying the impact of future wear on system health in a non-linear form, a smooth transition between historical data and predicted data is achieved, so that the health score can intuitively and comprehensively reflect the overall health status of the system, providing a clear quantitative basis for maintenance decisions.

[0061] Using the updated predicted life depreciation sequence to recalculate the overall life score, then generating the corresponding predicted wear score, and finally converting it into a health score not only ensures a smooth transition between the data prediction result and the health assessment, but also makes the prediction result have a clear quantitative threshold meaning, providing a direct indicator for decision-making.

[0062] Step 503: Set threshold Ⅰ and threshold Ⅱ, and threshold Ⅰ is less than threshold Ⅱ. Analyze according to the health score, and the logic is as follows: When the health score is greater than or equal to threshold Ⅱ, it is determined that the system is healthy, indicating that the next task is satisfied; When the health score is less than threshold II and greater than threshold I, it is determined that the system health level is relatively low, and a warning is issued, indicating that the next task can be satisfied but the vulnerable parts need to be replaced after the task ends; When the health score is less than or equal to threshold I, it is determined that there are potential hazards in the system, and an alarm is issued, indicating that it is insufficient to complete the next task.

[0063] According to the set threshold I and threshold II, the health score is classified and judged, realizing a complete closed-loop from data quantification to state evaluation, and then to warning and alarm. The classification and determination method not only provides timely alarm feedback for the subsequent operation safety of the equipment at a deep level, but also provides a clear operation guide for the overall maintenance strategy, enabling a strong connection and closed-loop feedback to be formed from the prediction model to the output of practical decision-making among all steps.

[0064] The above formulas are all dimensionless and take their numerical calculations. The formula is a formula obtained by software simulation of a large amount of collected data to approximate the real situation as much as possible. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0065] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0066] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0067] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A building electromechanical system health monitoring method based on digital twin technology, characterized in that: The specific steps include: Step 1: Obtain the replacement time of vulnerable parts of the air-conditioning system, obtain the working data of the air-conditioning system and the working data timestamp. The working data includes the operating power and operating time of each task. Compare the replacement time of vulnerable parts with the timestamp of the working data, and obtain the sample set of tasks experienced by each vulnerable part after each replacement. Step 2: Obtain the unit life loss rate of each vulnerable part each time it is replaced based on the sample set of tasks experienced by each vulnerable part after each replacement, and take the unit life loss rate of each vulnerable part in all replacement processes as the equivalent life loss rate of the vulnerable part. Sum up the equivalent life loss rates of all vulnerable parts and build a life loss digital twin model. Step 3: Obtain the last replacement time of each vulnerable component through the maintenance log, and the work data since the replacement, input the work data into the life loss digital twin model, obtain the life loss rate of each vulnerable component for each task, and form the life loss sequence of each vulnerable component. Combine the life loss sequence of each vulnerable component and the number of vulnerable components with high loss to construct an overall life score; Step 4: Based on the parameters and historical data of the air conditioning system, select the tasks whose operating power exceeds the rated power and their operating time and the tasks whose operating power is less than 70% of the rated power, obtain the average power of these tasks, and construct the system fatigue coefficient; Step 5: According to the next temperature control task to be performed by the air-conditioning system, the next operating power and the next operating time are obtained, and the life depreciation digital twin model is input to obtain the next life depreciation rate of each vulnerable component respectively. The corrected next life depreciation rate is obtained by correcting the system fatigue coefficient, and the corrected next life depreciation rate is added to the life depreciation sequence of each vulnerable part to form a predicted life depreciation sequence. The predicted wear score is obtained again based on the overall life score to generate a health score, and the state of the air-conditioning system is judged based on the relationship between the health score and thresholds I and II.

2. The building electromechanical system health monitoring method based on digital twin technology according to claim 1 is characterized by: The replacement time of vulnerable parts of the air-conditioning system is obtained. The vulnerable parts include air filters, fan bearings, fan belts, compressor valves, and condenser tubes. The replacement time is obtained through the maintenance log. The working data and working data timestamp of the air-conditioning system are obtained. The working data and timestamp are obtained through the working log. The working data include the operating power and operating time of each task. The operating power is the operating power of the entire air-conditioning system. The working data is mapped to the replacement time. The logic of the mapping is: Compare the replacement time with the timestamp of the working data to obtain the total number of tasks experienced by each vulnerable component after each replacement, as well as the operating power and operating time of each task, where the operating power of each vulnerable component is the operating power of the entire system during this task, and the operating time of each vulnerable component is the total operating time of the entire system during this task; Each vulnerable part is numbered, and a sample set of tasks experienced by each vulnerable part after each replacement is obtained, based on the following formula: ， in, Indicates Consumable parts After the replacement of the sample set, Indicates The operating power of the task, Indicates The running time of the task, Retrieve variables for consumable part numbers, , , Retrieve variables for task times, , , For this vulnerable parts The total number of tasks after replacement, Retrieve the variable for the number of changes, , , For the Total replacement times of wearing parts.

3. The building electromechanical system health monitoring method based on digital twin technology according to claim 2 is characterized in that: According to the sample set after each replacement of each vulnerable part, the equivalent life depreciation rate of each vulnerable part is obtained respectively. The logic is as follows: The rated life of each vulnerable part is obtained each time it is replaced. The rated life is the working time of the vulnerable part at the rated power. It is obtained from the manufacturer of the replaced vulnerable part. The unit life depreciation rate of each vulnerable part is obtained each time it is replaced. The unit life depreciation rate of each replacement represents the life depreciation rate of the unit working efficiency and unit working time of the vulnerable part from installation to replacement. The formula is as follows: ， in, Indicates Consumable parts Unit life loss rate after replacement, Indicates Consumable parts After the first replacement The operating power of the task, Indicates Consumable parts After the first replacement The running time of the task, Indicates The rated life of each vulnerable component, Retrieve variables for task times, , , For this vulnerable parts The total number of tasks after replacement, Retrieve the variable for the number of changes, , , For the Total replacement times of vulnerable parts, Retrieve variables for consumable part numbers, , ; Take the unit life loss rate of each vulnerable part during all replacement processes as the equivalent life loss rate of the vulnerable part, based on the following formula: ， in, For the The equivalent life loss rate of each vulnerable component is Indicates Consumable parts Equivalent life loss rate after replacement, Retrieve variables for consumable part numbers, , , Retrieve the variable for the number of changes, , , For the Total number of replacements of vulnerable parts; The equivalent life depreciation rate of each vulnerable component is summarized and a life depreciation digital twin model is constructed.

4. The building electromechanical system health monitoring method based on digital twin technology according to claim 3 is characterized by: The last replacement time of each vulnerable part is obtained through the maintenance log, and the working data of the vulnerable part since its replacement is obtained through the work log. The working data is input into the life loss digital twin model to obtain the life loss rate of each vulnerable part for each task, and form a life loss sequence. The formula is as follows: ， in, Indicates The life degradation sequence of vulnerable parts, Retrieve variables for consumable part numbers, , , Indicates The life loss rate of the task, Retrieve variables for task times, , , Indicates the number of tasks experienced.

5. The building electromechanical system health monitoring method based on digital twin technology according to claim 4 is characterized in that: According to the life loss sequence of all vulnerable parts, the overall life score is constructed according to the following formula: ， in, Score overall life expectancy, Indicates Consumable parts The life loss rate of the task, Retrieve variables for consumable part numbers, , , Retrieve variables for task times, , , Indicates the number of tasks experienced. Indicates High depreciation index of vulnerable parts, is the coefficient; Obtain the high loss judgment threshold. The logic for obtaining the high loss index is as follows: ， in, Indicates High depreciation index of vulnerable parts, is the high loss judgment threshold, Retrieve variables for task times, , , Indicates the number of runs experienced. Retrieve variables for consumable part numbers, , .

6. The building electromechanical system health monitoring method based on digital twin technology according to claim 5 is characterized in that: The air conditioning system parameters are obtained, and the system parameters of the air conditioning system include the design life of the air conditioning system, the average number of replacements of wearing parts and the rated power. The historical data of the air conditioning system is obtained, and the historical data includes the operating time and the total number of replacements. The operating time is the sum of the operating time of all tasks recorded in the work log, and the total number of replacements is the sum of the number of replacements of wearing parts of the air conditioning system. The number of replacements of wearing parts is obtained through the maintenance log. Tasks whose operating power exceeds the rated power and their operating time are screened and numbered. Tasks whose operating power is lower than 70% of the rated power are screened and the average power of these tasks is obtained. The system fatigue coefficient is constructed based on the following formula: ， in, is the fatigue coefficient, For the running time, For the design life, To replace the sum, is the average number of replacement times for vulnerable parts, Indicates The operating power of a task exceeding the rated power, is the rated power, For the The running time of the task with over-rated power, Retrieve variables for over-rated power tasks, , , The number of tasks that exceed the rated power, is the average power, is the coefficient.

7. The building electromechanical system health monitoring method based on digital twin technology according to claim 6 is characterized by: The next temperature control task of the air-conditioning system is obtained, and the next operating power and next operating time required by the air-conditioning system are obtained according to the temperature control task. The operating power and operating time are manually set according to the performance of the air-conditioning system. The next operating efficiency and the next operating time are input into the life loss digital twin model, and the next life loss rate of each vulnerable component is obtained respectively. The next life loss rate is corrected to obtain the corrected next life loss rate. The formula is as follows: ， in, is the corrected next life loss rate, is the next life loss rate, is the fatigue coefficient; The corrected next life loss rate is added to the life loss sequence to form a predicted life loss sequence, based on the following formula: ， in, Indicates The predicted life loss sequence of vulnerable parts, Indicates The life degradation sequence of vulnerable parts, Retrieve variables for consumable part numbers, , , is the corrected next life loss rate.

8. The building electromechanical system health monitoring method based on digital twin technology according to claim 7 is characterized in that: According to the predicted life loss sequence, the overall life score is formed again according to step 3, and the overall life score formed this time is calibrated as the predicted wear score to generate the health score. The formula is as follows: ， in, Score your health. Score overall life expectancy, Score for predicted wear.

9. The building electromechanical system health monitoring method based on digital twin technology according to claim 7 is characterized in that: Thresholds I and II are set, and threshold I is less than threshold II. Analysis is performed based on the health score. The logic is as follows: When the health score is greater than or equal to threshold II, the system is considered healthy, indicating that the next task is met; When the health score is less than threshold II and greater than threshold I, the system health level is considered low and an early warning is issued, indicating that the next task can be met but vulnerable parts need to be replaced after the task is completed; When the health score is less than or equal to threshold I, the system is deemed to have hidden dangers and an alarm is issued, indicating that it is not sufficient to complete the next task.

Images