A method and system for triggering a revisit based on building operation and maintenance data

Abstract

The present application relates to a kind of based on building operation data's revisit trigger method and system, the method includes collecting operation original data and preprocessing construction uniform evaluation dataset;Build building level, professional subsystem level, regional level, four layers of evaluation object of equipment level, distinguish revisit evaluation object and bottom layer evidence object, establish object attribution and generate record full-dimensional data's calculation index table;In turn calculate aging risk, operation abnormal risk, energy consumption abnormal risk, work order degradation risk, update opportunity index, the weighted contribution value of health risk index is calculated by the weighted fusion of each type of risk, to determine the main trigger cause of revisit;Pre-set multiple trigger threshold, when revisit evaluation object meets corresponding condition, automatically generate revisit work order and revisit notice.The present application relies on multi-level object to collect full operation data, multi-dimensional quantification equipment and building overall risk, automatically determine the need for revisit, realize building operation revisit intelligent, standardized determination.

Description

Technical Field

[0001] This invention relates to the fields of intelligent building operation and maintenance, building facility status evaluation and operation and maintenance big data processing, and in particular to a method and system for triggering follow-up visits based on building operation and maintenance data. Background Technology

[0002] With the popularization of the concept of building lifecycle management, intelligent operation and maintenance systems have been widely used in the building operation and maintenance phase. Existing intelligent operation and maintenance systems typically have the following functions: collecting building electromechanical equipment operation data, energy consumption data, and environmental parameters through IoT sensors; and recording fault handling and maintenance records through work order management modules, thereby achieving visualization and traceability of building operation status.

[0003] However, most existing technologies remain in a passive response closed loop of fault discovery and reactive dispatching, and their value is mainly reflected in reducing failure rate and energy consumption, failing to deeply mine the value of the massive amount of operation and maintenance data collected. Summary of the Invention

[0004] This invention provides a method and system for triggering follow-up visits based on building operation and maintenance data to solve the above-mentioned technical problems.

[0005] To address the aforementioned technical problems, this invention provides a method for triggering follow-up visits based on building operation and maintenance data, comprising the following steps:

[0006] Step 1: Collect raw operation and maintenance data and perform preprocessing to construct a unified evaluation dataset. The unified evaluation dataset includes object identification data, operation data, work order data, ledger and update cycle data, contact data, evaluation rules and trigger parameter data.

[0007] Step 2: Establish evaluation objects at the building level, professional subsystem level, regional level, and equipment level. Among them, the building level, professional subsystem level, and regional level evaluation objects are the objects of follow-up evaluation, and the equipment level evaluation objects are the underlying evidence objects. Establish the affiliation relationship of evaluation objects at each level and generate a calculation index table. The calculation index table records the evaluation object's level, the building it belongs to, the professional subsystem it belongs to, the region it belongs to, the set of subordinate equipment, the associated points, the operation data, the energy consumption data, the work order data, the ledger parameters, and the rule parameters.

[0008] Step 3: Using the evaluation period t as the calculation unit, calculate multiple risk indicators for each revisited evaluation object i. The multiple risk indicators include aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and upgrading and renovation opportunity index. The multiple risk indicators are weighted and integrated to obtain the health risk index, and the weighted contribution value of each type of risk is calculated. The main triggering reason for the revisit is determined based on the weighted contribution value.

[0009] Step 4: Preset trigger thresholds, which include health risk threshold, energy consumption risk threshold, operational anomaly risk threshold, work order risk threshold, recurring failure rate threshold, upgrade and renovation opportunity threshold, and aging equipment ratio threshold; if the return visit evaluation object i meets the judgment conditions corresponding to at least one type of trigger threshold, then the return visit evaluation object is determined to meet the return visit trigger requirements.

[0010] Step 5: Generate a follow-up work order and follow-up notification for the follow-up evaluation objects that meet the follow-up trigger requirements.

[0011] Preferably, in step 1, the preprocessing includes data deduplication, time format standardization, abnormal timestamp correction, filtering of obvious erroneous sensor values, marking equipment as offline, removing data from known downtime periods and data from construction periods, to obtain standardized evaluation data.

[0012] Preferably, in step 2, the attribution relationship of the evaluation objects includes at least the following: equipment-level evaluation objects belong to professional subsystem-level evaluation objects and are installed on regional-level evaluation objects; professional subsystem-level evaluation objects and regional-level evaluation objects belong to building-level objects; sensor points are associated with equipment-level evaluation objects, regional-level evaluation objects, or professional subsystem-level evaluation objects according to their installation location and monitoring objects; work orders are collected into equipment-level evaluation objects, professional subsystem-level evaluation objects, regional-level evaluation objects, or building-level evaluation objects according to the associated object number.

[0013] Preferably, in step 3, the calculation steps for the aging risk include: for a single piece of equipment d, calculating the proximity of the equipment's age based on the equipment's installation year, design life, and evaluation cycle; calculating the proportion of high-load operation time of equipment d within the current evaluation cycle t based on the equipment's load conditions; calculating the aging risk of a single piece of equipment under the current evaluation cycle t based on the proximity of the equipment's age and the proportion of high-load operation time; summarizing the aging risks of the equipment under the revisited evaluation object to obtain the aging risk of the revisited evaluation object; and statistically analyzing the proportion of aging equipment under the revisited evaluation object.

[0014] Preferably, in step 3, the calculation steps for the operational anomaly risk include: using historical data from the same period or a preset operational benchmark as the comparison object, calculating the degree of deviation of operational indicator m and the proportion of abnormal duration within the current evaluation period t; converting the degree of deviation and the proportion of abnormal duration into normalized scores respectively; and combining the deviation score weight and duration score weight of operational indicator m to calculate the operational anomaly risk of the revisited evaluation object i under the current evaluation period t.

[0015] Preferably, in step 3, the calculation steps for the energy consumption anomaly risk include: collecting the periodic electricity consumption of the revisited evaluation object within the evaluation period t, calculating the energy consumption per unit area and the year-on-year change rate of energy consumption; performing piecewise linear mapping on the energy consumption per unit area and the year-on-year change rate of energy consumption to obtain a normalized score; and combining the preset fusion weights of the two indicators, energy consumption per unit area and the year-on-year change rate of energy consumption, to calculate the overall energy consumption anomaly risk of the revisited evaluation object.

[0016] Preferably, in step 3, the calculation steps for the work order deterioration risk include: counting the number of work orders within the current evaluation period t, calculating four indicators: work order growth rate, repeat failure rate, average repair time, and proportion of serious work orders; converting each of the four indicators into a normalized score by comparing it with a preset threshold range; configuring the fusion weights corresponding to the four indicators, and weighted fusion to obtain the overall work order deterioration risk of the revisited evaluation object.

[0017] Preferably, in step 3, the calculation steps of the renovation and upgrading opportunity index include: calculating the running time of the revisited evaluation object from the start time to the current evaluation period t, comparing it with the standard renovation and upgrading period to obtain the renovation and upgrading period proximity, and obtaining the renovation and upgrading expiration index after limiting the range; based on the renovation and upgrading expiration index, adding aging risk, abnormal operation risk, abnormal energy consumption risk, and work order deterioration risk for correction, and limiting the corrected value to the range of 0 to 1 to obtain the renovation and upgrading opportunity index of the revisited evaluation object.

[0018] Preferably, in step 3, the calculation steps of the health risk index are as follows: For aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and renovation opportunity index, a fusion weight of not less than 0 and a total of 1 is assigned; each type of risk indicator is multiplied by its own weight to obtain the corresponding weighted contribution value; all weighted contribution values ​​are added together to obtain the health risk index of the revisited evaluation object; the weighted contribution values ​​are compared, and the risk with the highest value is used as the main triggering reason for the revisit, while risks with contribution values ​​exceeding a preset threshold are simultaneously used as auxiliary triggering reasons.

[0019] This invention also provides a follow-up visit triggering system based on building operation and maintenance data, used to execute the above-mentioned follow-up visit triggering method, including:

[0020] The data acquisition module is used to collect raw operation and maintenance data, perform preprocessing, and build a unified evaluation dataset;

[0021] The hierarchical object modeling module is used to establish four levels of evaluation objects, construct the hierarchical relationships of evaluation objects at each level, and generate a calculation index table;

[0022] The multi-risk quantification calculation module is used to calculate aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and renovation and upgrading opportunity index respectively. The weighted fusion is used to obtain the health risk index and output the weighted contribution value of each type of risk.

[0023] The follow-up visit trigger determination module is used to store preset trigger thresholds and iterate through each follow-up visit evaluation object to complete the follow-up visit trigger determination.

[0024] The work order push module is used to generate follow-up work orders and follow-up notifications, and push the follow-up notifications to the corresponding contacts.

[0025] Compared with existing technologies, the revisit triggering method and system based on building operation and maintenance data provided by this invention have the following advantages:

[0026] 1. This invention constructs four levels of evaluation objects: building level, professional subsystem level, regional level, and equipment level. It relies on the attribution relationship and calculation index table to hierarchically collect scattered operation and maintenance data, realize unified management of multi-dimensional data, and fully explore the value of operation and maintenance data assets such as operation, energy consumption, work orders, and ledgers.

[0027] 2. This invention conducts a multi-dimensional quantitative assessment of aging risk, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, and upgrade and renovation opportunity index. The health risk index is obtained by weighted fusion and the primary and secondary triggering causes are distinguished. The risk assessment is comprehensive and objective, and the follow-up judgment standard is uniformly quantified to avoid the bias of subjective human judgment.

[0028] 3. This invention automatically triggers follow-up work orders by setting multiple risk thresholds, eliminating the need for manual inspection of equipment status one by one. It realizes automatic identification of building operation and maintenance follow-up needs and automatic generation of work orders, promoting the operation and maintenance mode from reactive post-event maintenance to proactive pre-event follow-up and prediction, and reducing operation and maintenance labor costs.

[0029] 4. This invention utilizes a hierarchical assessment architecture to output risk results for equipment, regions, professional subsystems, and the entire building, accurately locating high-risk equipment, clearly defining follow-up targets, and improving the pertinence of operation and maintenance renovation and inspection work, as well as the standardization of operation and maintenance management. Attached Figure Description

[0030] Figure 1 This is a flowchart of a follow-up triggering method based on building operation and maintenance data in a specific embodiment of the present invention. Detailed Implementation

[0031] To illustrate the technical solutions of the invention in more detail, specific embodiments are listed below to demonstrate the technical effects; it should be emphasized that these embodiments are used to illustrate the invention and not to limit the scope of the invention.

[0032] The present invention provides a method for triggering follow-up visits based on building operation and maintenance data, such as... Figure 1 As shown, it includes the following steps:

[0033] Step 1: Collect and standardize operation and maintenance data

[0034] Step 1.1: The system collects building operation and maintenance related data to form a unified evaluation dataset. The unified evaluation dataset includes:

[0035] 1) Object identification data, including the construction unit, building, professional subsystem, region, equipment, sensor location, and work order number;

[0036] 2) Operational data, including temperature, humidity, air quality, water consumption, power consumption, equipment start / stop status, running time, instantaneous voltage, instantaneous current, instantaneous power, and meter readings;

[0037] 3) Work order data, including work order type, creation time, completion time, processing time, associated objects, fault type, and severity level;

[0038] 4) Ledger and renovation cycle data, including building type, building area, year of building commissioning, equipment type, year of equipment installation, equipment design life, rated power, rated capacity, equipment ownership, equipment importance, standard renovation cycle and the time of the most recent renovation;

[0039] 5) Contact data, including information on the construction unit's contact person and the person in charge of the operating department;

[0040] 6) Evaluation rules and triggering parameter data, including evaluation cycle, types of indicators participating in the evaluation, operating baseline data, load rate threshold, anomaly judgment threshold, scoring mapping parameters, indicator weight, risk weight, risk level threshold, threshold required for follow-up trigger judgment, fault statistics rules, rules for determining the main triggering causes, and the value of a very small positive number ε used to prevent the denominator from being zero.

[0041] Step 1.2: The system preprocesses the above data, including data deduplication, time format standardization, abnormal timestamp correction, filtering of obvious erroneous sensor values, marking offline equipment, removing data from known downtime periods and data from construction periods, to obtain standardized evaluation data.

[0042] Step 2: Establish a hierarchical evaluation object model

[0043] Step 2.1: Establish evaluation objects at the building level, professional subsystem level, regional level, and equipment level. Among them, building-level objects, professional subsystem-level objects, and regional-level objects serve as follow-up evaluation objects to determine whether a follow-up request should be generated; equipment-level objects serve as underlying evidence objects to provide risk evidence such as aging, failure, energy consumption, and operational anomalies, but do not directly generate follow-up request.

[0044] Step 2.2: Establish the attribution relationships between evaluation objects at each level. Specifically, equipment-level evaluation objects belong to professional subsystem-level evaluation objects and are installed within regional-level evaluation objects; professional subsystem-level and regional-level evaluation objects belong to building-level objects; sensor locations are associated with equipment-level, regional-level, or professional subsystem-level evaluation objects according to their installation location and monitored objects; work orders are aggregated to equipment-level, professional subsystem-level, regional-level, or building-level evaluation objects according to their associated object numbers.

[0045] Step 2.3: Establish a calculation index table for each evaluation object based on the above attribution relationships. The calculation index table records the evaluation object's level, building, professional subsystem, region, subordinate equipment set, associated locations, operational data, energy consumption data, work order data, ledger parameters, and rule parameters. It is used to determine the data range, risk collection objects, and evaluation result recording locations for subsequent risk calculations.

[0046] Step 3: Calculate the health risk index of the subjects in the follow-up evaluation.

[0047] The system calculates health risk indices for building-level, professional subsystem-level, and regional-level follow-up evaluation objects. The health risk index is derived by integrating aging risk, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, and renovation / upgrade opportunity indices.

[0048] Specifically, the system uses the evaluation period t as the calculation unit to calculate multiple risk indicators for each revisited evaluation object. For ease of subsequent explanation, several frequently used variables are defined here: i represents the building-level, professional subsystem-level, or regional-level revisited evaluation object; d represents the equipment-level underlying evidence object, hereinafter referred to as equipment d; m represents the type of operational indicator participating in the evaluation; τ represents the sampling time within the evaluation period t; t-1 represents the previous evaluation period; t-12 represents the historical concurrent evaluation period corresponding to the current evaluation period; DeviceSet(i) represents the set of equipment-level evidence objects under the revisited evaluation object i; MetricSet(i) represents the set of operational indicators of the revisited evaluation object i participating in the operational anomaly evaluation; T_total represents the total duration of the evaluation period; ε represents a very small positive number used to prevent the denominator from being zero.

[0049] In addition, except for variables explicitly labeled t, t-1, t-12, or τ in the formula, all risk results in this step that do not explicitly label the evaluation period subscript represent the results calculated for the corresponding object under the current evaluation period t. For example, AgingRisk(i), OperationRisk(i), EnergyRisk(i), WorkOrderRisk(i), RenovationOpportunityIndex(i), HealthRiskIndex(i), and AgedDeviceRate(i) are all the corresponding calculation results for the revisited evaluation object i under the current evaluation period t.

[0050] Step 3.1: Calculate aging risk

[0051] Step 3.1.1: Calculate the similarity in equipment age:

[0052] For equipment-level evidence, the system calculates the similarity of equipment lifespan based on the equipment's installation year, design life, and evaluation period:

[0053] AgeRatio(d)=(CurrentYear-InstallYear(d)) / Life(d)

[0054] This formula indicates how close the equipment's service life is to its design life. Where CurrentYear is the year of the current evaluation period, InstallYear(d) is the year equipment d was installed, Life(d) is the design life of equipment d, and AgeRatio(d) is the closeness of equipment d's service life. The larger the AgeRatio(d), the closer the equipment is to or beyond its design life.

[0055] Step 3.1.2: Calculate the percentage of equipment operating under high load.

[0056] Further, based on the equipment load conditions, calculate the percentage of high-load operating time for equipment d within the current evaluation period t:

[0057] LoadRate(d,τ)=InstantPower(d,τ) / RatedPower(d);

[0058] OverloadRate(d,t)=T(LoadRate(d,τ)>LoadThreshold(d)) / T_total;

[0059] This set of formulas is used to introduce the high-load operation factor of equipment. In this formula, InstantPower(d,τ) is the instantaneous power of equipment d at time τ, RatedPower(d) is the rated power of equipment d, LoadThreshold(d) is the high-load judgment threshold of equipment d, T(LoadRate(d,τ)>LoadThreshold(d)) represents the cumulative duration for which the load rate exceeds LoadThreshold(d) within the evaluation period t, and OverloadRate(d,t) is the proportion of high-load operation time of equipment d within the current evaluation period t, used to reflect the amplifying effect of long-term high-load operation on the risk of equipment aging.

[0060] Step 3.1.3: Calculate equipment aging risk

[0061] AgingRisk(d)=min(1,AgeRatio(d)×(1+α×OverloadRate(d,t)));

[0062] Where α is the load correction coefficient, and AgingRisk(d) is the aging risk of equipment d under the current evaluation period t. This formula combines the proximity of equipment age and the proportion of high-load operation to obtain the aging risk of a single piece of equipment. min(1,·) is used to limit the risk value to the range of 0 to 1, which facilitates unified integration with other risk indicators.

[0063] Step 3.1.4: Summarize the equipment aging risks to the evaluation target i.

[0064] For regional, professional subsystem, and building-level follow-up evaluation targets, the system summarizes the aging risks of their subordinate equipment to obtain the aging risk of the evaluation target:

[0065] AgingRisk(i)=Σ[λ(i,d)×AgingRisk(d)],d∈DeviceSet(i);

[0066] Wherein, λ(i,d) represents the aggregate weight of device d in the revisited evaluation object i, and satisfies Σλ(i,d)=1. λ(i,d) represents the importance of different devices to the evaluation object, reflecting the higher impact of key devices, large-capacity devices, or devices with large service areas on the overall aging risk; AgingRisk(i) represents the overall aging risk of the revisited evaluation object i under the current evaluation period t. The aggregate weight is determined based on at least one of device importance, rated capacity, service area, and device quantity ratio. For scenarios where no device importance or capacity parameters are configured, each subordinate device can be aggregated with equal weight.

[0067] Step 3.1.5: Calculate the proportion of aging equipment

[0068] The system also calculates the percentage of aging equipment under the evaluation target i based on the equipment's aging risk:

[0069] AgedDeviceCount(i) = Count{d∈DeviceSet(i)|AgingRisk(d)≥equipment aging risk threshold};

[0070] AgedDeviceRate(i)=AgedDeviceCount(i) / [TotalDeviceCount(i)+ε];

[0071] The equipment aging risk threshold is derived from the evaluation rules and triggering parameter data in step 1.1; AgedDeviceCount(i) is the number of devices in DeviceSet(i) whose AgingRisk(d) reaches the equipment aging risk threshold; TotalDeviceCount(i) is the total number of devices in DeviceSet(i); AgedDeviceRate(i) is the proportion of aging devices in the current evaluation period t for the revisited evaluation object i; Count{·} indicates counting the devices that meet the conditions in parentheses. AgedDeviceRate can be used as an independent triggering condition to identify revisit scenarios for upgrades and renovations where "although the overall risk of a single item may not be the highest, the proportion of aging devices is relatively high".

[0072] Step 3.2: Calculate the risk of operational anomalies

[0073] The system calculates the risk of operational anomalies based on operating indicators such as temperature, humidity, air quality, water consumption, power consumption, equipment operating time, start-up and shutdown status, instantaneous voltage, instantaneous current and instantaneous power.

[0074] The system uses historical data from the same period or a preset operating benchmark as a comparison object to calculate the degree of deviation and duration of abnormality of operating indicators within the current evaluation period t:

[0075] Deviation(i,m)=|Actual(i,m)-Baseline(i,m)| / (Baseline(i,m)+ε);

[0076] AbnormalDurationRate(i,m)=T_abnormal(i,m) / T_total;

[0077] OperationRisk(i,m)=w1(m)×DeviationScore(i,m)+w2(m)×DurationScore(i,m);

[0078] The degree of deviation reflects the severity of the anomaly, while the duration reflects its stability or longevity. Combining both can prevent false triggers caused by short-term fluctuations and can also identify long-term minor anomalies. Specifically, Actual(i,m) is the actual value of the performance indicator m of the evaluated object i within the current evaluation period t; Baseline(i,m) is the historical baseline value or preset baseline value of the performance indicator m; Deviation(i,m) is the degree of deviation of the performance indicator m in the current evaluation period t; T_abnormal(i,m) is the cumulative duration for which the performance indicator m is judged as abnormal within the current evaluation period t; AbnormalDurationRate(i,m) is the percentage of abnormal duration for the performance indicator m within the current evaluation period t; and Deviation... onScore(i,m) is the normalized deviation score obtained after the score mapping of Deviation(i,m); DurationScore(i,m) is the normalized duration score obtained after the score mapping of AbnormalDurationRate(i,m); w1(m) and w2(m) are the deviation score weight and duration score weight of the operation indicator m, and satisfy w1(m)+w2(m)=1; OperationRisk(i,m) is the operation abnormality risk of the revisited evaluation object i under the operation indicator m in the current evaluation period t.

[0079] OperationRisk(i)=Σ[μ(i,m)×OperationRisk(i,m)],m∈MetricSet(i);

[0080] μ(i,m) represents the weight of operational indicator m in the evaluation object i, and satisfies Σμ(i,m)=1. μ(i,m) indicates the importance of different operational indicators in a specific evaluation object. For example, indicators such as air quality, temperature and humidity, voltage, current, and power can be configured with different weights according to the building type or professional system type. OperationRisk(i) represents the operational anomaly risk of the evaluation object i under the current evaluation period t.

[0081] Step 3.3: Calculate the risk of abnormal energy consumption

[0082] The system calculates the risk of energy consumption anomalies within the current evaluation period t based on meter readings and energy consumption statistics. For building-level, regional-level, or professional subsystem-level follow-up evaluations, the evaluation can be based on periodic electricity consumption, energy consumption per unit area, and year-on-year change rate.

[0083] When energy consumption data comes from equipment-level evidence objects or sub-metering points, the system summarizes the energy consumption data of the equipment-level evidence objects or metering points under the evaluation object i according to the object affiliation relationship, and obtains the cumulative meter reading or periodic electricity consumption of the evaluation object i.

[0084] ElectricityConsumption(i,t)=MeterReading(i,t)-MeterReading(i,t-1);

[0085] EUI(i,t)=ElectricityConsumption(i,t) / Area(i);

[0086] EnergyYoY(i,t)=[ElectricityConsumption(i,t)-ElectricityConsumption(i,t-12)] / [ElectricityConsumption(i,t-12)+ε];

[0087] In this set of formulas, MeterReading(i,t) represents the cumulative meter reading of the evaluated object i at the end of the current evaluation period t; MeterReading(i,t-1) represents the cumulative meter reading of the evaluated object i at the end of the previous evaluation period; ElectricityConsumption(i,t) represents the periodic electricity consumption of the evaluated object i within the current evaluation period t; Area(i) represents the building area, metered area, or service area corresponding to the evaluated object i; EUI(i,t) represents the energy consumption per unit area of ​​the evaluated object i in the current evaluation period t; and EnergyYoY(i,t) represents the year-on-year change rate of energy consumption of the evaluated object i in the current evaluation period t. This set of formulas characterizes energy consumption anomalies from three perspectives: periodic electricity consumption, energy consumption per unit area, and year-on-year change rate. EUI is used to measure the energy consumption level per unit area horizontally, while EnergyYoY is used to identify abnormal growth compared to the same period in previous years vertically.

[0088] The system maps energy consumption per unit area and year-on-year energy consumption change rate to normalized scores respectively:

[0089] EUI_score(i,t)=min(1,max(0,[EUI(i,t)-EUI_low(i)] / [EUI_high(i)-EUI_low(i)]))

[0090] YoY_score(i,t)=min(1,max(0,[EnergyYoY(i,t)-YoY_low(i)] / [YoY_high(i)-YoY_low(i)]))

[0091] Wherein, EUI_low(i) is the low-risk threshold for energy consumption per unit area of ​​the evaluated object i, and EUI_high(i) is the high-risk threshold for energy consumption per unit area of ​​the evaluated object i; YoY_low(i) is the low-risk threshold for year-on-year growth in energy consumption of the evaluated object i, and YoY_high(i) is the high-risk threshold for year-on-year growth in energy consumption of the evaluated object i. This formula uses a piecewise linear mapping between low-risk and high-risk thresholds to convert energy consumption indicators of different dimensions into risk scores between 0 and 1. Below the low-risk threshold, the risk approaches 0; above the high-risk threshold, the risk approaches 1.

[0092] EnergyRisk(i)=a1×EUI_score(i,t)+a2×YoY_score(i,t)

[0093] Wherein, EUI_score(i,t) is the normalized score obtained by threshold piecewise linear mapping of EUI(i,t); YoY_score(i,t) is the normalized score obtained by threshold piecewise linear mapping of EnergyYoY(i,t); a1 and a2 are the fusion weights of energy consumption anomaly risk, and both a1 and a2 are not less than 0, and satisfy a1+a2=1. If more attention is paid to energy consumption intensity, a1 is increased; if more attention is paid to abnormal growth trend, a2 is increased; EnergyRisk(i) is the energy consumption anomaly risk of the revisited evaluation object i under the current evaluation period t.

[0094] Step 3.4: Calculate the risk of work order degradation

[0095] The system calculates the work order degradation risk based on the number of work orders within the current evaluation period t, the work order growth rate, the recurring failure rate, the mean time to repair, and the proportion of serious work orders.

[0096] WorkOrderGrowthRate(i,t)=[OrderCount(i,t)-OrderCount(i,t-1)] / [OrderCount(i,t-1)+ε];

[0097] RepeatFaultRate(i,t)=RepeatFaultCount(i,t) / [TotalFaultCount(i,t)+ε];

[0098] MTTR(i,t)=SumRepairDuration(i,t) / [RepairCount(i,t)+ε];

[0099] SevereOrderRate(i,t)=SevereOrderCount(i,t) / [OrderCount(i,t)+ε];

[0100] Wherein, OrderCount(i,t) is the number of work orders collected by the follow-up evaluation object i in the current evaluation period t; OrderCount(i,t-1) is the number of work orders collected by the follow-up evaluation object i in the previous evaluation period; WorkOrderGrowthRate(i,t) is the work order growth rate of the follow-up evaluation object i in the current evaluation period t; RepeatFaultCount(i,t) is the number of repeated faults of the follow-up evaluation object i in the current evaluation period t; TotalFaultCount(i,t) is the total number of fault-type work orders of the follow-up evaluation object i in the current evaluation period t; RepeatFaultRate(i,t) is... The formulas evaluate the recurrence rate of work order failures for evaluation object i in the current evaluation period t; SumRepairDuration(i,t) is the sum of processing times for maintenance work orders for evaluation object i in the current evaluation period t; RepairCount(i,t) is the number of maintenance work orders for evaluation object i in the current evaluation period t; MTTR(i,t) is the average repair time for evaluation object i in the current evaluation period t; SevereOrderCount(i,t) is the number of work orders for evaluation object i that reach the preset severity threshold in the current evaluation period t; and SevereOrderRate(i,t) is the percentage of severe work orders for evaluation object i in the current evaluation period t. This set of formulas evaluates the risk of work order degradation from four dimensions: changes in the number of work orders, recurrence failures, maintenance efficiency, and severity. It can reflect whether the building or system operation and maintenance status has evolved from an occasional problem to a persistent problem.

[0101] The system maps the work order growth rate, repeat failure rate, average repair time, and proportion of critical work orders into normalized scores:

[0102] GrowthScore(i,t)=min(1,max(0,[WorkOrderGrowthRate(i,t)-GrowthLow(i)] / [GrowthHigh(i)-GrowthLow(i)]));

[0103] RepeatFaultScore(i,t)=min(1,max(0,[RepeatFaultRate(i,t)-RepeatFaultLow(i)] / [RepeatFaultHigh(i)-RepeatFaultLow(i)]));

[0104] MTTRScore(i,t)=min(1,max(0,[MTTR(i,t)-MTTRLow(i)] / [MTTRHigh(i)-MTTRLow(i)]));

[0105] SeverityScore(i,t)=min(1,max(0,[SevereOrderRate(i,t)-SeverityLow(i)] / [SeverityHigh(i)-SeverityLow(i)]));

[0106] Wherein, GrowthLow(i) is the low-risk threshold for the work order growth rate of the feedback evaluation object i, and GrowthHigh(i) is the high-risk threshold for the work order growth rate of the feedback evaluation object i, and GrowthHigh(i) is greater than GrowthLow(i); RepeatFaultLow(i) is the low-risk threshold for the repeated failure rate of the feedback evaluation object i, and RepeatFaultHigh(i) is the high-risk threshold for the repeated failure rate of the feedback evaluation object i, and RepeatFaultHigh(i) is greater than RepeatFaultLow(i). ultLow(i); MTRLow(i) is the low-risk threshold for the average repair time of the back-visit evaluation object i, and MTRHigh(i) is the high-risk threshold for the average repair time of the back-visit evaluation object i, and MTRHigh(i) is greater than MTRLow(i); SeverityLow(i) is the low-risk threshold for the proportion of severe work orders of the back-visit evaluation object i, and SeverityHigh(i) is the high-risk threshold for the proportion of severe work orders of the back-visit evaluation object i, and SeverityHigh(i) is greater than SeverityLow(i).

[0107] This set of formulas converts work order indicators of different calibers into a unified risk score of 0 to 1, enabling work order growth rate, repeat failure rate, average repair time and the proportion of serious work orders to be integrated in the same risk model.

[0108] WorkOrderRisk(i)=b1×GrowthScore(i,t)+b2×RepeatFaultScore(i,t)+b3×MTTRScore(i,t)+b4×SeverityScore(i,t);

[0109] Among them, GrowthScore(i,t), RepeatFaultScore(i,t), MTTRScore(i,t), and SeverityScore(i,t) are normalized scores obtained by threshold-based piecewise linear mapping of work order growth rate, repeat failure rate, mean time to repair, and proportion of severe work orders, respectively. The values ​​range from 0 to 1, with higher values ​​indicating higher work order degradation risk. b1 to b4 are the fusion weights of work order degradation risk, corresponding to work order growth, repeat failure, mean time to repair, and proportion of severe work orders, respectively. b1, b2, b3, and b4 are all not less than 0 and satisfy b1+b2+b3+b4=1. The specific values ​​can be configured according to the actual operation and maintenance management priorities. WorkOrderRisk(i) is the work order degradation risk of the revisited evaluation object i under the current evaluation period t.

[0110] Step 3.5: Calculate the opportunity index for renovation and upgrading.

[0111] The system calculates the upgrade and renovation opportunity index based on the proximity of the evaluated object to the standard upgrade and renovation cycle in the current evaluation period t, combined with the risks of abnormal energy consumption, abnormal operation, work order deterioration, and aging.

[0112] ElapsedRenovationTime(i,t)=EvaluationTime(t)-LastRenovationTime(i);

[0113] RenovationCycleRatio(i,t)=ElapsedRenovationTime(i,t) / StandardRenovationCycle(i);

[0114] RenovationDueIndex(i,t)=min(1,max(0,RenovationCycleRatio(i,t)));

[0115] Where EvaluationTime(t) is the evaluation time corresponding to the current evaluation period; LastRenovationTime(i) is the most recent renovation and upgrade completion time of the revisited evaluation object i; when the revisited evaluation object i does not have a most recent renovation and upgrade completion time, the building commissioning time, the professional subsystem activation time, the area activation time, or the installation time of its subordinate main equipment shall be used as the starting time; ElapsedRenovationTime(i,t) is the running time of the revisited evaluation object i from the starting time to the current evaluation period t.

[0116] This set of formulas is used to calculate how close the evaluation object is to the standard renovation and upgrading cycle. The higher the RenovationDueIndex, the closer the object is to or beyond the standard renovation and upgrading cycle, and the more suitable it is to be included in the renovation and upgrading follow-up assessment.

[0117] RenovationOpportunityIndex(i)=min(1,RenovationDueIndex(i,t)×[1+q1×EnergyRisk(i)+q2×OperationRisk(i)+q3×WorkOrderRisk(i)+q4×AgingRisk(i)])

[0118] Wherein, StandardRenovationCycle(i) is the standard renovation cycle corresponding to the revisited evaluation object i, and StandardRenovationCycle(i) is greater than 0; ElapsedRenovationTime(i,t) and StandardRenovationCycle(i) use the same time unit. RenovationCycleRatio(i,t) is the renovation cycle proximity; RenovationDueIndex(i,t) is the renovation due date index, with a value ranging from 0 to 1. The higher the value, the closer the revisited evaluation object i is to or beyond the standard renovation cycle.

[0119] EnergyRisk(i) is the energy consumption anomaly risk obtained in step 3.3; OperationRisk(i) is the operation anomaly risk obtained in step 3.2; WorkOrderRisk(i) is the work order deterioration risk obtained in step 3.4; and AgingRisk(i) is the aging risk obtained in step 3.1. EnergyRisk(i), OperationRisk(i), WorkOrderRisk(i), and AgingRisk(i) are all normalized risk values, ranging from 0 to 1.

[0120] q1 to q4 are the correction coefficients for the upgrade and renovation opportunity index based on the risks of abnormal energy consumption, abnormal operation, work order deterioration, and aging, respectively, and q1, q2, q3, and q4 are all not less than 0. These correction coefficients are used to improve the upgrade and renovation opportunity index based on actual operational deterioration, abnormal energy consumption, work order deterioration, and equipment aging; q1 + q2 + q3 + q4 does not need to equal 1.

[0121] RenovationOpportunityIndex(i) is the opportunity index for renovation and upgrading of the revisited evaluation object i, with a value ranging from 0 to 1. The higher the value, the more likely the revisited evaluation object i is to have renovation, energy-saving renovation, system upgrade or revisit development value.

[0122] This formula does not simply judge the opportunity for renovation based on the number of years. Instead, it adds actual operational evidence such as abnormal energy consumption, abnormal operation, work order deterioration, and equipment aging to the basis of the proximity of the renovation cycle, thereby improving the accuracy of the business triggered by the follow-up visit.

[0123] Step 3.6: Integrate current health risk indices

[0124] The system weights and integrates the aging risk, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, and upgrade / renovation opportunity index under the current evaluation period t to obtain the current health risk index of the revisited evaluation object:

[0125] HealthRiskIndex(i)=p1×AgingRisk(i)+p2×OperationRisk(i)+p3×EnergyRisk(i)+p4×WorkOrderRisk(i)+p5×RenovationOpportunityIndex(i)

[0126] Wherein, AgingRisk(i), OperationRisk(i), EnergyRisk(i), WorkOrderRisk(i) and RenovationOpportunityIndex(i) are the sub-risks or indices under the current evaluation period t obtained in steps 3.1 to 3.5, respectively.

[0127] This formula is the core fusion formula of this method, which unifies aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and upgrade and renovation opportunity index into a health risk index, which is used to determine whether a follow-up visit is triggered.

[0128] p1 to p5 are the combined weights of the health risk index, where p1 is the aging risk weight, p2 is the operational anomaly risk weight, p3 is the energy consumption anomaly risk weight, p4 is the work order deterioration risk weight, and p5 is the upgrade and renovation opportunity index weight. p1, p2, p3, p4, and p5 are all not less than 0 and satisfy the following conditions:

[0129] p1+p2+p3+p4+p5=1;

[0130] This constraint is used to ensure that the health risk index remains within the normalized range of 0 to 1, facilitating horizontal comparisons and unified threshold judgments between different buildings, different professional subsystems, and different regions.

[0131] Since AgingRisk(i), OperationRisk(i), EnergyRisk(i), WorkOrderRisk(i), and RenovationOpportunityIndex(i) are all normalized values ​​and range from 0 to 1, HealthRiskIndex(i) ranges from 0 to 1, provided that p1 to p5 are all not less than 0 and the sum of their weights is 1. The higher the HealthRiskIndex(i) value, the more likely the revisited evaluation object i is to have repair, energy-saving renovation, service upgrade, or revisit value.

[0132] The system determines the primary triggering cause based on the weighted contribution value of each sub-risk to the health risk index within the current evaluation period t. The weighted contribution value includes:

[0133] AgingContribution(i)=p1×AgingRisk(i);

[0134] OperationContribution(i)=p2×OperationRisk(i);

[0135] EnergyContribution(i)=p3×EnergyRisk(i);

[0136] WorkOrderContribution(i)=p4×WorkOrderRisk(i);

[0137] RenovationContribution(i)=p5×RenovationOpportunityIndex(i);

[0138] Among them, AgingContribution(i) is the contribution value of aging risk; OperationContribution(i) is the contribution value of operation anomaly risk; EnergyContribution(i) is the contribution value of energy consumption anomaly risk; WorkOrderContribution(i) is the contribution value of work order deterioration risk; and RenovationContribution(i) is the contribution value of renovation and upgrading opportunities. All of the above contribution values ​​are weighted contribution values ​​under the current evaluation period t.

[0139] The system identifies the highest weighted contribution value among the above as the primary triggering cause. When multiple weighted contribution values ​​reach a preset contribution threshold, the system records all causes simultaneously as primary triggering causes, which are used to generate the cause explanation in the follow-up work order. In other words, the system not only determines whether a follow-up visit is needed, but also identifies the primary triggering cause based on the weighted contribution value of each sub-risk, making the generated follow-up work order interpretable. The preset contribution threshold is derived from the evaluation rules and triggering parameter data or the primary triggering cause determination rules in step 1.1.

[0140] Step 4: Determine the triggering requirement for a follow-up visit

[0141] Based on the health risk index, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, recurring failure rate, upgrade and renovation opportunity index, and aging equipment ratio obtained in step 3, the system determines whether a follow-up visit trigger requirement should be generated for the follow-up evaluation object i in the current evaluation period t. When the follow-up evaluation object i meets any of the following conditions, the system determines that the follow-up evaluation object i has met the follow-up visit trigger requirement in the current evaluation period t, as shown in Table 1:

[0142] Table 1

[0143] Serial Number Triggering conditions Follow-up visit type 1 HealthRiskIndex(i) ≥ Health Risk Threshold Repair and renovation follow-up visit 2 EnergyRisk(i) ≥ Energy Risk Threshold Energy-saving renovation follow-up visit 3 OperationRisk(i) ≥ Operational anomaly risk threshold Fault troubleshooting follow-up 4 WorkOrderRisk(i) ≥ WorkOrder Risk Threshold Follow-up visit to confirm the necessity of renovation 5 RepeatFaultRate(i,t) ≥ Repeat Fault Rate Threshold Fault troubleshooting follow-up 6 RenovationOpportunityIndex(i) ≥ Renovation Opportunity Threshold Follow-up visits for renovation and upgrading opportunities 7 AgedDeviceRate(i) ≥ AgedDeviceRate Threshold Follow-up visits for renovation and upgrading opportunities

[0144] Wherein, HealthRiskIndex(i), EnergyRisk(i), OperationRisk(i), WorkOrderRisk(i), RenovationOpportunityIndex(i), and AgedDeviceRate(i) are respectively the health risk index, energy consumption abnormality risk, operation abnormality risk, work order deterioration risk, renovation and upgrading opportunity index, and aging equipment ratio of the revisited evaluation object i under the current evaluation period t obtained in step 3; RepeatFaultRate(i,t) is the repeat failure rate of the revisited evaluation object i under the current evaluation period t obtained in step 3.4.

[0145] The aforementioned health risk threshold, energy consumption risk threshold, operational anomaly risk threshold, work order risk threshold, repeat failure rate threshold, upgrade and renovation opportunity threshold, and aging equipment proportion threshold are all derived from the evaluation rules and trigger parameter data in step 1.1.

[0146] When the same subject of a follow-up evaluation meets multiple triggering conditions, the system generates a follow-up work order and records all the conditions met, the corresponding follow-up type, the main triggering reason, the lower-level subjects with high contribution, and a list of key equipment evidence.

[0147] Step 5: Send SMS push and generate follow-up work order

[0148] The system generates a follow-up notification and a follow-up work order for the follow-up evaluation objects that meet the follow-up trigger requirements, and pushes the follow-up notification to relevant personnel via SMS based on the contact person data and business department head data in step 1.

[0149] This invention also provides a follow-up visit triggering system based on building operation and maintenance data, used to execute the above-mentioned follow-up visit triggering method, including:

[0150] The data acquisition module is used to perform step 1: collect raw operation and maintenance data and perform preprocessing to build a unified evaluation dataset;

[0151] The hierarchical object modeling module is used to perform step 2: establish four-level evaluation objects, construct the hierarchical relationship of each level of evaluation objects, and generate a calculation index table;

[0152] The multi-risk quantification calculation module is used to perform step 3: calculate aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and renovation opportunity index respectively, and obtain the health risk index by weighted fusion and output the weighted contribution value of each type of risk;

[0153] The follow-up visit trigger determination module is used to execute step 4: store the preset trigger threshold and traverse each follow-up visit evaluation object to complete the follow-up visit trigger determination.

[0154] The work order push module is used to perform step 5: generate a follow-up work order and follow-up notification, and push (such as SMS) the follow-up notification to the corresponding contact person.

[0155] The following detailed explanation uses the air conditioning subsystem of an office building A as the subject of the follow-up evaluation.

[0156] Office building A, designated B001, has a floor area of ​​12,000 square meters and was put into operation in 2012. The air conditioning subsystem, designated S-HVAC, serves the office areas on floors one through ten of office building A. The current evaluation period t is from 00:00 on May 1, 2026 to 24:00 on May 31, 2026, with a total evaluation period T_total of 744 hours. The previous evaluation period t-1 was in April 2026, and the historical evaluation period t-12 was in May 2025. The smallest positive number ε, used to prevent the denominator from being zero, is set to 0.000001.

[0157] S1: Collect and standardize operation and maintenance data

[0158] S1.1: The system collects building operation and maintenance related data to form a unified evaluation dataset.

[0159] In this embodiment, the system collects operation and maintenance related data of office building A and its air conditioning subsystem S-HVAC to form a unified evaluation dataset.

[0160] The unified evaluation dataset includes object identification data, operation data, work order data, ledger and update / renovation cycle data, contact person data, and evaluation rules and trigger parameter data.

[0161] The object identification data includes: construction unit number CU001, building number B001, professional subsystem number S-HVAC, area number R01 to R10, equipment number d1 to d4, sensor location number P001 to P030, and work order number WO-202605-001 to WO-202605-018.

[0162] The equipment-level evidence objects under the S-HVAC subsystem include chiller unit d1, cooling tower fan d2, combined air conditioning unit d3, and chilled water pump d4. The ledger data for each piece of equipment is shown in Table 2.

[0163] Table 2

[0164] Equipment Number Equipment Name Year of installation Design life Rated power Summarize the weights λ(i,d) d1 chiller unit 2013 15 years 220kW 0.45 d2 Cooling tower fan 2015 12 years 15kW 0.15 d3 Combined air conditioning units 2018 10 years 45kW 0.25 d4 chilled water pump 2021 10 years 30kW 0.15

[0165] The operating data includes: instantaneous power of chiller units, instantaneous power of cooling tower fans, instantaneous power of combined air conditioning units, instantaneous power of chilled water pumps, chilled water supply temperature, indoor temperature of office areas, equipment start-up and shutdown status, equipment running time, instantaneous voltage, instantaneous current, and instantaneous power.

[0166] Work order data includes: maintenance work orders, fault work orders, and complaint work orders related to the Air Conditioning Subsystem S-HVAC and its subordinate equipment d1 to d4 within the current evaluation period t. The number of work orders collected for the Air Conditioning Subsystem S-HVAC within the current evaluation period t is 18, and the number of work orders collected for the Air Conditioning Subsystem S-HVAC within the previous evaluation period t-1 is 10.

[0167] The ledger and renovation cycle data include: Office Building A is an office building with a building area of ​​12,000 square meters and was put into operation in 2012; the most recent renovation of the air conditioning subsystem S-HVAC was completed on January 1, 2016, and the standard renovation cycle is 10 years.

[0168] The contact information includes: the mobile phone number of contact person A from the construction unit, and the mobile phone number of person B in charge of the operations department.

[0169] The evaluation rules and triggering parameters include: the evaluation period is monthly; the high load judgment threshold LoadThreshold(d) is 0.80; the load correction coefficient α is 0.30; the equipment aging risk threshold is 0.80; the health risk threshold is 0.750; the energy consumption risk threshold is 0.650; the operation abnormality risk threshold is 0.700; the work order risk threshold is 0.700; the repeat failure rate threshold is 0.300; the upgrade and renovation opportunity threshold is 0.800; and the aging equipment proportion threshold is 0.500.

[0170] In this embodiment, the fusion weights of the health risk index are set as follows: aging risk weight p1 is 0.20, abnormal operation risk weight p2 is 0.25, abnormal energy consumption risk weight p3 is 0.20, work order deterioration risk weight p4 is 0.20, and the renewal and renovation opportunity index weight p5 is 0.15, and p1+p2+p3+p4+p5=1.

[0171] S1.2: The system preprocesses the above data to obtain standardized evaluation data.

[0172] The system preprocesses the data collected by S1.1.

[0173] Specifically, the system deletes duplicate sensor data uploads; converts the time field uploaded by different devices to a unified format of "year-month-day T hour:minute:second"; corrects abnormal timestamps that are significantly outside the evaluation period; removes obviously erroneous sensor values ​​where the chilled water supply temperature is below 0℃ or above 40℃; marks periods when sensors have not uploaded data for 60 consecutive minutes as offline periods; removes data from known downtime maintenance periods; and removes abnormal energy consumption data caused by partial construction in office building A.

[0174] After the above preprocessing, the system obtains standardized evaluation data for the current evaluation period t, which serves as the data basis for subsequently establishing a hierarchical evaluation object model, calculating the health risk index, and determining the need for follow-up visits.

[0175] S2: Establish a hierarchical evaluation object model

[0176] S2.1: Establish evaluation objects at the building level, professional subsystem level, regional level, and equipment level.

[0177] The system establishes building-level object B001, professional subsystem-level object S-HVAC, area-level objects R01 to R10, and equipment-level objects d1 to d4.

[0178] Among them, building-level object B001, professional subsystem-level object S-HVAC, and regional-level objects R01 to R10 are return visit evaluation objects, which can be used to determine whether a return visit requirement is generated; equipment-level objects d1 to d4 are underlying evidence objects, used to provide evidence of aging, failure, energy consumption, and abnormal operation, but equipment-level objects themselves do not directly generate return visit requirements.

[0179] In this embodiment, the system selects the professional subsystem-level object S-HVAC as the current return visit evaluation object i.

[0180] S2.2: Establish ownership relationships between objects

[0181] The system establishes the hierarchical relationships between equipment-level objects, professional subsystem-level objects, regional-level objects, and building-level objects.

[0182] In this embodiment, chiller unit d1, cooling tower fan d2, combined air conditioning unit d3, and chilled water pump d4 all belong to the air conditioning subsystem S-HVAC; the air conditioning subsystem S-HVAC belongs to office building A, i.e., building-level object B001; combined air conditioning unit d3 serves areas R01 to R10; chilled water supply temperature sensor point P001 is associated with the air conditioning subsystem S-HVAC; office area indoor temperature sensor points P010 to P019 are respectively associated with areas R01 to R10; fault work orders related to the chiller unit, cooling tower fan, combined air conditioning unit, and chilled water pump are respectively associated with the corresponding equipment and further aggregated to the air conditioning subsystem S-HVAC.

[0183] S2.3: Establish a calculation index table for each evaluation object based on the above object relationships.

[0184] Based on the object relationships established in S2.2, the system creates a calculation index table for the air conditioning subsystem S-HVAC.

[0185] The calculated index table records the following information:

[0186] Follow-up evaluation object number: S-HVAC;

[0187] Object level: Professional subsystem level;

[0188] Building: B001;

[0189] The subordinate device set DeviceSet(i) is: {d1, d2, d3, d4};

[0190] The set of operational indicators involved in the evaluation of operational anomalies is MetricSet(i): chilled water supply temperature, office area indoor temperature, and equipment operating time;

[0191] Related points: P001 to P030;

[0192] Related operating data: temperature, start / stop status, running time, instantaneous power, instantaneous current, and instantaneous voltage;

[0193] Related energy consumption data: S-HVAC sub-meter readings;

[0194] Related work order data: WO-202605-001 to WO-202605-018;

[0195] Related ledger parameters: equipment installation year, design life, rated power, standard upgrade and renovation cycle, and the date of the most recent upgrade and renovation;

[0196] Association rule parameters: load rate threshold, anomaly detection threshold, scoring mapping parameters, risk weight, and return visit trigger threshold.

[0197] The system determines the data range, risk aggregation objects, and evaluation result recording location for subsequent risk calculations based on the calculation index table.

[0198] S3: Calculate the health risk index of the subjects being evaluated in the follow-up visit.

[0199] The system calculates a health risk index for S-HVAC systems, which are subject to follow-up evaluation at the professional subsystem level. The health risk index is derived by integrating aging risk, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, and upgrade / renovation opportunity index.

[0200] In this embodiment, i represents the S-HVAC professional subsystem; d represents the equipment-level evidence object; m represents the type of operational indicator participating in the evaluation; τ represents the sampling time within the evaluation period t; DeviceSet(i) is {d1, d2, d3, d4}; MetricSet(i) is {chilled water supply temperature, office area indoor temperature, equipment running time}; T_total is 744 hours.

[0201] S3.1: Calculate aging risk

[0202] S3.1.1: Computational equipment age proximity

[0203] The system calculates the similarity of equipment lifespan based on the year of equipment installation, design life, and the year of the current evaluation period.

[0204] The current evaluation period is for the year 2026.

[0205] The approximation of the service life of chiller unit d1 is as follows:

[0206] AgeRatio(d1)=(2026-2013) / 15=0.867.

[0207] The approximation of the service life of cooling tower fan d2 is as follows:

[0208] AgeRatio(d2)=(2026-2015) / 12=0.917.

[0209] The approximation of the service life of the combined air conditioning unit d3 is as follows:

[0210] AgeRatio(d3)=(2026-2018) / 10=0.800.

[0211] The approximate lifespan of chilled water pump d4 is as follows:

[0212] AgeRatio(d4)=(2026-2021) / 10=0.500.

[0213] S3.1.2: Calculate the percentage of equipment operating under high load.

[0214] The system calculates the equipment load rate based on the equipment's instantaneous power and rated power, and counts the cumulative duration during which the equipment load rate exceeds the high load judgment threshold of 0.80 within the evaluation period t.

[0215] Within the current evaluation period t, the system statistics yielded the following data, as shown in Table 3:

[0216] Table 3

[0217] Equipment Number High load running time Total evaluation period T_total High load operation rate d1 96 hours 744 hours 0.129 d2 40 hours 744 hours 0.054 d3 88 hours 744 hours 0.118 d4 20 hours 744 hours 0.027

[0218] Therefore, the high-load operation ratio of chiller unit d1 is 96 / 744=0.129; the high-load operation ratio of cooling tower fan d2 is 40 / 744=0.054; the high-load operation ratio of combined air conditioning unit d3 is 88 / 744=0.118; and the high-load operation ratio of chilled water pump d4 is 20 / 744=0.027.

[0219] S3.1.3: Calculate the risk of equipment aging

[0220] The system calculates equipment aging risk based on the similarity of equipment age and the proportion of high-load operation. The load correction factor α is set to 0.30.

[0221] The aging risk of chiller unit d1 is:

[0222] AgingRisk(d1)

[0223] =min(1,0.867×(1+0.30×0.129))

[0224] =0.900.

[0225] The aging risk of cooling tower fan d2 is as follows:

[0226] AgingRisk(d2)

[0227] =min(1,0.917×(1+0.30×0.054))

[0228] =0.931.

[0229] The aging risk of the combined air conditioning unit d3 is as follows:

[0230] AgingRisk(d3)

[0231] =min(1,0.800×(1+0.30×0.118))

[0232] =0.828.

[0233] The aging risk of chilled water pump d4 is as follows:

[0234] AgingRisk(d4)

[0235] =min(1,0.500×(1+0.30×0.027))

[0236] =0.504.

[0237] S3.1.4: Summarize the equipment aging risks in the follow-up evaluation objects i

[0238] The system aggregates equipment aging risks into the air conditioning subsystem S-HVAC according to the equipment aggregation weight λ(i,d).

[0239] Where λ(i,d1)=0.45, λ(i,d2)=0.15, λ(i,d3)=0.25, λ(i,d4)=0.15, and the sum of the aggregate weights is 1.

[0240] AgingRisk(i)

[0241] =0.45×0.900+0.15×0.931+0.25×0.828+0.15×0.504

[0242] =0.828.

[0243] Therefore, the overall aging risk of the S-HVAC professional subsystem under the current evaluation period t is 0.828.

[0244] S3.1.5: Calculate the proportion of aging equipment

[0245] The system uses an equipment aging risk threshold of 0.80 to count the number of aging devices in the S-HVAC subsystem.

[0246] Since AgingRisk(d1) = 0.900, AgingRisk(d2) = 0.931, and AgingRisk(d3) = 0.828, all of which are not less than the equipment aging risk threshold of 0.80, d1, d2, and d3 are identified as aging equipment. The AgingRisk(d4) of chilled water pump d4 is 0.504, which is less than the equipment aging risk threshold of 0.80; therefore, d4 is not identified as aging equipment.

[0247] AgedDeviceCount(i) = 3.

[0248] TotalDeviceCount(i) = 4.

[0249] AgedDeviceRate(i)

[0250] =3 / (4+ε)

[0251] =0.750.

[0252] Therefore, the proportion of aging equipment in the S-HVAC professional subsystem under the current evaluation period t is 0.750.

[0253] S3.2: Calculate the risk of operational anomalies

[0254] The system calculates the risk of operational anomalies based on the chilled water supply temperature, the indoor temperature of the office area, and the operating time of the equipment.

[0255] In this embodiment, the system uses historical data from the same period as the operating benchmark and scores the deviation and the duration of the anomaly. The deviation score weight w1 is set to 0.60, and the duration score weight w2 is set to 0.40.

[0256] Within the current evaluation period t, the system has collected the following operational indicator data, as shown in Table 4:

[0257] Table 4

[0258] Operating Indicator m Current actual value Actual(i,m) Historical benchmark (i,m) Duration of abnormality chilled water supply temperature 9.2℃ 7.0℃ 168 hours Office indoor temperature 27.4℃ 25.5℃ 132 hours Equipment runtime 510 hours 430 hours 110 hours

[0259] For chilled water supply temperature, the degree of deviation is:

[0260] Deviation(i,m1)

[0261] =|9.2-7.0| / (7.0+ε)

[0262] =0.314.

[0263] The system maps a deviation of 0.314 to a normalized deviation score of 1.000. The percentage of abnormal duration is as follows:

[0264] AbnormalDurationRate(i,m1)

[0265] =168 / 744

[0266] =0.226.

[0267] The system mapped the percentage of abnormal duration (0.226) to a normalized duration score of 0.879.

[0268] therefore:

[0269] OperationRisk(i,m1)

[0270] =0.60×1.000+0.40×0.879

[0271] =0.952.

[0272] Regarding the indoor temperature in the office area, the degree of deviation is:

[0273] Deviation(i,m2)

[0274] =|27.4-25.5| / (25.5+ε)

[0275] =0.075.

[0276] The system maps a deviation of 0.075 to a normalized deviation score of 0.681. The percentage of abnormal duration is as follows:

[0277] AbnormalDurationRate(i,m2)

[0278] =132 / 744

[0279] =0.177.

[0280] The system maps the percentage of abnormal duration (0.177) to a normalized duration score of 0.637.

[0281] therefore:

[0282] OperationRisk(i,m2)

[0283] =0.60×0.681+0.40×0.637

[0284] =0.663.

[0285] Regarding device runtime, the degree of deviation is:

[0286] Deviation(i,m3)

[0287] =|510-430| / (430+ε)

[0288] =0.186.

[0289] The system maps a deviation of 0.186 to a normalized deviation score of 0.544. The percentage of abnormal duration is as follows:

[0290] AbnormalDurationRate(i,m3)

[0291] =110 / 744

[0292] =0.148.

[0293] The system maps the percentage of abnormal duration (0.148) to a normalized duration score of 0.489.

[0294] therefore:

[0295] OperationRisk(i,m3)

[0296] =0.60×0.544+0.40×0.489

[0297] =0.522.

[0298] In this embodiment, the weights of the indicators for chilled water supply temperature, office area indoor temperature, and equipment operating time are 0.40, 0.35, and 0.25, respectively.

[0299] Therefore, the operational risks of the S-HVAC subsystem are as follows:

[0300] OperationRisk(i)

[0301] =0.40×0.952+0.35×0.663+0.25×0.522

[0302] =0.743.

[0303] S3.3: Calculate the risk of abnormal energy consumption

[0304] The system calculates the risk of abnormal energy consumption based on the sub-meter readings and energy consumption statistics of the S-HVAC professional subsystem.

[0305] In this embodiment, the service area Area(i) of the S-HVAC professional subsystem is 12,000 square meters. The cumulative meter reading MeterReading(i,t) at the end of the current evaluation period t is 1,263,000 kWh, and the cumulative meter reading MeterReading(i,t-1) at the end of the previous evaluation period t-1 is 1,015,000 kWh.

[0306] Therefore, the periodic electricity consumption within the current evaluation period t is:

[0307] ElectricityConsumption(i,t)

[0308] =1263000-1015000

[0309] =248000kWh.

[0310] Energy consumption per unit area is:

[0311] EUI(i,t)

[0312] =248000 / 12000

[0313] =20.667 kWh / ㎡.

[0314] In this embodiment, the low-risk threshold EUI_low(i) for energy consumption per unit area is 15 kWh / m², and the high-risk threshold EUI_high(i) for energy consumption per unit area is 25 kWh / m².

[0315] therefore:

[0316] EUI_score(i,t)

[0317] =min(1,max(0,(20.667-15) / (25-15)))

[0318] =0.567.

[0319] The system reads the historical electricity consumption (ElectricityConsumption(i,t-12)) for the same evaluation period t-12 as 205000 kWh. Therefore, the year-on-year change rate of energy consumption is:

[0320] EnergyYoY(i,t)

[0321] =(248000-205000) / (205000+ε)

[0322] =0.210.

[0323] In this embodiment, the low-risk threshold YoY_low(i) for year-on-year energy consumption growth is 0.05, and the high-risk threshold YoY_high(i) for year-on-year energy consumption growth is 0.25.

[0324] therefore:

[0325] YoY_score(i,t)

[0326] =min(1,max(0,(0.210-0.05) / (0.25-0.05)))

[0327] =0.799.

[0328] In this embodiment, the energy consumption score weight a1 per unit area is 0.50, and the energy consumption year-on-year score weight a2 is 0.50.

[0329] therefore:

[0330] EnergyRisk(i)

[0331] =0.50×0.567+0.50×0.799

[0332] =0.683.

[0333] Therefore, the energy consumption anomaly risk of the S-HVAC professional subsystem in the current evaluation period t is 0.683.

[0334] S3.4: Calculate the risk of work order degradation

[0335] The system calculates the work order degradation risk of the S-HVAC subsystem based on the number of work orders, work order growth rate, recurring failure rate, average repair time, and proportion of serious work orders within the current evaluation period t.

[0336] During the current evaluation period t, the number of work orders collected by the S-HVAC professional subsystem is 18 (OrderCount(i,t)); during the previous evaluation period t-1, the number of work orders collected was 10 (OrderCount(i,t-1)).

[0337] Therefore, the work order growth rate is:

[0338] WorkOrderGrowthRate(i,t)

[0339] =(18-10) / (10+ε)

[0340] =0.800.

[0341] In this embodiment, the low-risk threshold for work order growth rate, GrowthLow(i), is 0.10, and the high-risk threshold for work order growth rate, GrowthHigh(i), is 0.80.

[0342] therefore:

[0343] GrowthScore(i,t)

[0344] =min(1,max(0,(0.800-0.10) / (0.80-0.10)))

[0345] =1.000.

[0346] Within the current evaluation period t, the total number of fault-related work orders (TotalFaultCount(i,t)) is 15, and the number of repeated faults (RepeatFaultCount(i,t)) is 6.

[0347] therefore:

[0348] RepeatFaultRate(i,t)

[0349] =6 / (15+ε)

[0350] =0.400.

[0351] In this embodiment, the repeat fault low risk threshold RepeatFaultLow(i) is 0.10, and the repeat fault high risk threshold RepeatFaultHigh(i) is 0.50.

[0352] therefore:

[0353] RepeatFaultScore(i,t)

[0354] =min(1,max(0,(0.400-0.10) / (0.50-0.10)))

[0355] =0.750.

[0356] Within the current evaluation period t, the sum of the processing times for repair work orders, SumRepairDuration(i,t), is 120 hours, and the number of repair work orders, RepairCount(i,t), is 15.

[0357] therefore:

[0358] MTTR(i,t)

[0359] =120 / (15+ε)

[0360] =8,000 hours.

[0361] In this embodiment, the low-risk threshold MTRLow(i) for average repair time is 4 hours, and the high-risk threshold MTRHigh(i) for average repair time is 10 hours.

[0362] therefore:

[0363] MTTRScore(i,t)

[0364] =min(1,max(0,(8-4) / (10-4)))

[0365] =0.667.

[0366] Within the current evaluation period t, the number of work orders with a severity level reaching the preset severity threshold, SevereOrderCount(i,t), is 5, and the total number of work orders, OrderCount(i,t), is 18.

[0367] therefore:

[0368] SevereOrderRate(i,t)

[0369] =5 / (18+ε)

[0370] =0.278.

[0371] In this embodiment, the low-risk threshold SeverityLow(i) for the proportion of serious work orders is 0.10, and the high-risk threshold SeverityHigh(i) for the proportion of serious work orders is 0.30.

[0372] therefore:

[0373] SeverityScore(i,t)

[0374] =min(1,max(0,(0.278-0.10) / (0.30-0.10)))

[0375] =0.889.

[0376] In this embodiment, the weighting for the work order growth rate is b1, which is 0.25; the weighting for the repeat failure rate is b2, which is 0.25; the weighting for the average repair time is b3, which is 0.25; and the weighting for the proportion of serious work orders is b4, which is 0.25.

[0377] therefore:

[0378] WorkOrderRisk(i)

[0379] =0.25×1.000+0.25×0.750+0.25×0.667+0.25×0.889

[0380] =0.826.

[0381] Therefore, the work order degradation risk of the S-HVAC professional subsystem under the current evaluation period t is 0.826.

[0382] S3.5: Calculate the opportunity index for renovation and upgrading.

[0383] The system calculates the upgrade and renovation opportunity index based on the proximity of the S-HVAC subsystem to the standard upgrade and renovation cycle, and in combination with the risks of abnormal energy consumption, abnormal operation, work order deterioration, and aging.

[0384] In this embodiment, the last renovation and upgrade completion time of the S-HVAC professional subsystem is January 1, 2016, the current evaluation period corresponds to the evaluation time (T) on May 31, 2026, and the standard renovation and upgrade cycle (I) is 10 years.

[0385] The elapsedRenovationTime(i,t) from the completion of the most recent update to the current evaluation period is approximately 10.4 years.

[0386] therefore:

[0387] RenovationCycleRatio(i,t)

[0388] =10.4 / 10

[0389] =1.040.

[0390] RenovationDueIndex(i,t)

[0391] =min(1,max(0,1.040))

[0392] =1.000.

[0393] In this embodiment, the energy consumption anomaly risk correction coefficient q1 is 0.15, the operation anomaly risk correction coefficient q2 is 0.15, the work order deterioration risk correction coefficient q3 is 0.10, and the aging risk correction coefficient q4 is 0.10.

[0394] therefore:

[0395] RenovationOpportunityIndex(i)

[0396] =min(1,1.000×[1+0.15×0.683+0.15×0.743+0.10×0.826+0.10×0.828])

[0397] =1.000.

[0398] Therefore, the upgrade and renovation opportunity index for the S-HVAC professional subsystem under the current evaluation period t is 1.000.

[0399] S3.6: Integrate current health risk index

[0400] The system weights and integrates the aging risk, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, and upgrade and renovation opportunity index under the current evaluation period t to obtain the current health risk index of the air conditioning subsystem S-HVAC.

[0401] In this embodiment, AgingRisk(i)=0.828, OperationRisk(i)=0.743, EnergyRisk(i)=0.683, WorkOrderRisk(i)=0.826, RenovationOpportunityIndex(i)=1.000.

[0402] The aging risk weight p1 is set to 0.20, the abnormal operation risk weight p2 is set to 0.25, the abnormal energy consumption risk weight p3 is set to 0.20, the work order deterioration risk weight p4 is set to 0.20, and the renovation and upgrading opportunity index weight p5 is set to 0.15.

[0403] therefore:

[0404] HealthRiskIndex(i)

[0405] =0.20×0.828+0.25×0.743+0.20×0.683+0.20×0.826+0.15×1.000

[0406] =0.803.

[0407] Therefore, the health risk index of the S-HVAC professional subsystem is 0.803 in the current evaluation period t.

[0408] The system further calculates the weighted contribution of each sub-risk to the health risk index:

[0409] AgingContribution(i)

[0410] =0.20×0.828

[0411] =0.166.

[0412] OperationContribution(i)

[0413] =0.25×0.743

[0414] =0.186.

[0415] EnergyContribution(i)

[0416] =0.20×0.683

[0417] =0.137.

[0418] WorkOrderContribution(i)

[0419] =0.20×0.826

[0420] =0.165.

[0421] RenovationContribution(i)

[0422] =0.15×1.000

[0423] =0.150.

[0424] Among them, the contribution value of operational anomaly risk is 0.186, which is the highest contribution value. Therefore, the system identifies operational anomaly as the primary triggering cause. Since the contribution values ​​of aging risk (0.166), work order deterioration risk (0.165), and upgrade and renovation opportunity (0.150) all reach the preset contribution threshold of 0.150, the system also records equipment aging, work order deterioration, and upgrade and renovation opportunities as secondary triggering causes.

[0425] S4: Determine if a follow-up visit triggers a requirement.

[0426] Based on the health risk index, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, recurring failure rate, upgrade and renovation opportunity index, and aging equipment ratio obtained from S3, the system determines whether to generate a return visit trigger requirement for the S-HVAC object under the current evaluation period t.

[0427] In this embodiment, the system uses the following trigger thresholds, as shown in Table 5:

[0428] Table 5

[0429] Trigger indicators Trigger threshold Health risk threshold 0.750 Energy consumption risk threshold 0.650 Operational anomaly risk threshold 0.700 Work order risk threshold 0.700 Repeat failure rate threshold 0.300 Upgrade and renovation opportunity threshold 0.800 Aging equipment ratio threshold 0.500

[0430] Based on the calculation results of S3, the various indicators of the S-HVAC professional subsystem under the current evaluation period t are shown in Table 6:

[0431] Table 6

[0432] index Calculation results Trigger threshold Trigger? Corresponding follow-up type HealthRiskIndex(i) 0.803 0.750 yes Repair and renovation follow-up visit EnergyRisk(i) 0.683 0.650 yes Energy-saving renovation follow-up visit OperationRisk(i) 0.743 0.700 yes Fault troubleshooting follow-up WorkOrderRisk(i) 0.826 0.700 yes Follow-up visit to confirm the necessity of renovation RepeatFaultRate(i,t) 0.400 0.300 yes Fault troubleshooting follow-up RenovationOpportunityIndex(i) 1.000 0.800 yes Follow-up visits for renovation and upgrading opportunities AgedDeviceRate(i) 0.750 0.500 yes Follow-up visits for renovation and upgrading opportunities

[0433] As shown in the table above, the S-HVAC subsystem for air conditioning simultaneously meets multiple return visit triggering conditions during the current evaluation period t.

[0434] The system determines that the S-HVAC object being evaluated has met the requirements for a follow-up visit. Since the same object being evaluated has triggered multiple conditions, the system does not generate multiple follow-up work orders. Instead, it generates a single comprehensive follow-up work order, which records all the conditions met, the corresponding follow-up type, the main triggering reason, the secondary triggering reason, the lower-level objects with high contribution, and a list of evidence for key equipment.

[0435] In this embodiment, the primary triggering reason recorded by the system is operational abnormality, while secondary triggering reasons include equipment aging, work order deterioration, and opportunities for upgrades and modifications. The list of key equipment evidence recorded by the system includes:

[0436] 1. The aging risk of chiller unit d1 is 0.900, and the high-load operation time is 96 hours;

[0437] 2. The aging risk of cooling tower fan d2 is 0.931, and the high-load operation time is 40 hours;

[0438] The aging risk of the 3-unit combined air conditioning system d3 is 0.828, and the high-load operation time is 88 hours;

[0439] 4. The electricity consumption of the S-HVAC professional subsystem during the current evaluation period is 248,000 kWh, an increase of 21.0% compared to 205,000 kWh in the same period of the previous year;

[0440] 5. The number of work orders for the S-HVAC subsystem in the current evaluation period is 18, an increase of 80.0% compared to 10 in the previous evaluation period;

[0441] 6. The repeat failure rate of the S-HVAC professional subsystem is 40.0%, exceeding the repeat failure rate threshold of 30.0%.

[0442] S5: Send SMS push notification and generate follow-up work order

[0443] The system generates a return visit notification and a return visit work order for the S-HVAC subsystem that meets the return visit trigger requirements, and pushes the return visit notification to relevant personnel via SMS based on the contact person data and business department head data in S1.

[0444] In this embodiment, the system determines that the SMS push recipients include contact person A from the construction unit and head of the operations department B.

[0445] The content of the SMS notification generated by the system is as follows:

[0446] "

Building Operation and Maintenance Follow-up Reminder

[0447] The system also generates a follow-up work order. The follow-up work order includes the following information:

[0448] Follow-up work order number: RV-202605-SHVAC-001;

[0449] Related building number: B001;

[0450] Related feedback evaluation target: S-HVAC (Air Conditioning and Ventilation) subsystem;

[0451] Evaluation period: May 2026;

[0452] Types of follow-up visits: follow-up visits for repair and renovation, follow-up visits for energy-saving renovation, follow-up visits for troubleshooting, follow-up visits for the necessity of renovation, and follow-up visits for opportunities for upgrading and renovation;

[0453] Health risk index: 0.803;

[0454] Aging risk: 0.828;

[0455] Risk of operational anomalies: 0.743;

[0456] Energy consumption anomaly risk: 0.683;

[0457] Work order degradation risk: 0.826;

[0458] Upgrading and renovation opportunity index: 1.000;

[0459] Percentage of aging equipment: 0.750;

[0460] Repeat failure rate: 0.400;

[0461] Main triggering reason: runtime error;

[0462] Auxiliary triggering reasons: equipment aging, work order deterioration, and opportunities for upgrading and renovation;

[0463] Triggering conditions: Health risk index reaches the threshold, energy consumption abnormality risk reaches the threshold, operation abnormality risk reaches the threshold, work order deterioration risk reaches the threshold, repeated failure rate reaches the threshold, upgrade and renovation opportunity index reaches the threshold, and aging equipment ratio reaches the threshold.

[0464] Evidence of key equipment: chiller unit d1, cooling tower fan d2, and combined air conditioning unit d3;

[0465] Recommended actions: Arrange for operations personnel to contact the construction unit, organize maintenance personnel to conduct on-site inspections of the air conditioning system's operating status, and evaluate equipment upgrades, energy-saving renovations, and system optimization plans;

[0466] Target audience: Construction unit contact person A, business department head B;

[0467] Work order status: Pending follow-up.

[0468] Through the above steps, the system can calculate the aging risk, operational anomaly risk, energy consumption anomaly risk, work order deterioration risk, and renovation opportunity index of the S-HVAC subsystem based on building operation and maintenance data, and integrate these risks into a health risk index. When any follow-up visit trigger condition is met, the system automatically generates a follow-up visit work order and sends a follow-up visit notification to relevant personnel via SMS, thereby realizing the automatic identification and proactive triggering of follow-up visit needs based on building operation and maintenance data.

[0469] In summary, the follow-up visit triggering method and system based on building operation and maintenance data provided by this invention hierarchically collects various types of building operation and maintenance data and constructs a calculation index table. It quantifies risk indicators from multiple dimensions, weights and fuses them to obtain a health risk index, and automatically triggers follow-up visit work orders by combining multiple thresholds. This invention fully explores the value of operation and maintenance data, realizes standardized risk assessment and intelligent follow-up visit determination at all levels, and upgrades the operation and maintenance mode from passive maintenance to proactive predictive follow-up visits.

[0470] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for triggering a follow-up visit based on building operation and maintenance data, characterized in that, Includes the following steps: Step 1: Collect raw operation and maintenance data and perform preprocessing to construct a unified evaluation dataset. The unified evaluation dataset includes object identification data, operation data, work order data, ledger and update cycle data, contact data, evaluation rules and trigger parameter data. Step 2: Establish evaluation objects at the building level, professional subsystem level, regional level, and equipment level. Among them, the building level, professional subsystem level, and regional level evaluation objects are the objects of follow-up evaluation, and the equipment level evaluation objects are the objects of underlying evidence. Establish the attribution relationship of evaluation objects at all levels and generate a calculation index table. The calculation index table records the evaluation object's level, building, professional subsystem, region, subordinate equipment set, associated points, operation data, energy consumption data, work order data, ledger parameters and rule parameters. Step 3: Using the evaluation period t as the calculation unit, calculate multiple risk indicators for each revisited evaluation object i. The multiple risk indicators include aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and upgrading and renovation opportunity index. The multiple risk indicators are weighted and integrated to obtain the health risk index, and the weighted contribution value of each type of risk is calculated. The main triggering reason for the revisit is determined based on the weighted contribution value. Step 4: Preset trigger thresholds, which include health risk threshold, energy consumption risk threshold, operational anomaly risk threshold, work order risk threshold, recurring failure rate threshold, upgrade and renovation opportunity threshold, and aging equipment ratio threshold; if the return visit evaluation object i meets the judgment conditions corresponding to at least one type of trigger threshold, then the return visit evaluation object is determined to meet the return visit trigger requirements. Step 5: Generate a follow-up work order and follow-up notification for the follow-up evaluation objects that meet the follow-up trigger requirements.

2. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, In step 1, the preprocessing includes data deduplication, time format standardization, abnormal timestamp correction, filtering of obvious erroneous sensor values, marking equipment as offline, removing data from known downtime periods and data from construction periods, to obtain standardized evaluation data.

3. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, In step 2, the attribution relationship of the evaluation objects includes at least the following: equipment-level evaluation objects belong to professional subsystem-level evaluation objects and are installed on regional-level evaluation objects; professional subsystem-level evaluation objects and regional-level evaluation objects belong to building-level objects; sensor points are associated with equipment-level evaluation objects, regional-level evaluation objects, or professional subsystem-level evaluation objects according to their installation location and monitoring objects; work orders are collected into equipment-level evaluation objects, professional subsystem-level evaluation objects, regional-level evaluation objects, or building-level evaluation objects according to the associated object number.

4. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, Step 3, the calculation steps for the aging risk include: For a single piece of equipment d, the similarity of equipment age is calculated based on the equipment installation year, design life, and evaluation cycle; the proportion of high-load operation time of equipment d in the current evaluation cycle t is calculated based on the equipment load; the aging risk of a single piece of equipment in the current evaluation cycle t is calculated based on the similarity of equipment age and the proportion of high-load operation time; the aging risks of the equipment under the revisited evaluation object are summarized to obtain the aging risk of the revisited evaluation object; and the proportion of aging equipment under the revisited evaluation object is statistically analyzed.

5. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, Step 3, the calculation steps for the operational anomaly risk, include: Using historical data from the same period or a preset operating benchmark as a comparison object, calculate the degree of deviation of operating indicator m and the proportion of abnormal duration within the current evaluation period t; convert the degree of deviation and the proportion of abnormal duration into normalized scores respectively; combine the deviation score weight and duration score weight of operating indicator m to calculate the operational anomaly risk of the revisited evaluation object i under the current evaluation period t.

6. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, Step 3, the calculation steps for the abnormal energy consumption risk include: Collect the periodic electricity consumption of the evaluated objects within the evaluation period t, and calculate the energy consumption per unit area and the year-on-year change rate of energy consumption. Perform piecewise linear mapping on the energy consumption per unit area and the year-on-year change rate of energy consumption to obtain a normalized score. Combine the preset fusion weights of the two indicators, energy consumption per unit area and the year-on-year change rate of energy consumption, to calculate the overall energy consumption anomaly risk of the evaluated objects.

7. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, Step 3, the calculation steps for the work order deterioration risk include: The number of work orders within the current evaluation period t is counted separately, and four indicators are calculated: work order growth rate, repeat failure rate, average repair time, and proportion of serious work orders. Each of the four indicators is converted into a normalized score by comparing it with a preset threshold range. The fusion weights corresponding to the four indicators are configured, and the weighted fusion is used to obtain the overall work order deterioration risk of the revisited evaluation object.

8. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, Step 3, the calculation steps for the renovation and upgrading opportunity index include: The running time of the subject being evaluated from the start time to the current evaluation period t is calculated and compared with the standard renovation and upgrading period to obtain the closeness of the renovation and upgrading period. After limiting the range, the renovation and upgrading expiration index is obtained. Based on the renovation and upgrading expiration index, the aging risk, operation abnormality risk, energy consumption abnormality risk, and work order deterioration risk are added for correction. The corrected value is limited to the range of 0 to 1 to obtain the renovation and upgrading opportunity index of the subject being evaluated.

9. The method for triggering a follow-up visit based on building operation and maintenance data according to claim 1, characterized in that, In step 3, the calculation steps for the health risk index are as follows: For aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and upgrading and renovation opportunity index, respectively, a fusion weight of not less than 0 and a total weight of 1 is assigned; Each risk indicator is multiplied by its own weight to obtain the corresponding weighted contribution value; all weighted contribution values ​​are summed to obtain the health risk index of the subject of the follow-up evaluation; the weighted contribution values ​​are compared, and the risk with the highest value is used as the main trigger for the follow-up, while risks with contribution values ​​exceeding the preset threshold are used as auxiliary triggers.

10. A follow-up triggering system based on building operation and maintenance data, characterized in that, The method for executing the revisit triggering method according to any one of claims 1 to 9 includes: The data acquisition module is used to collect raw operation and maintenance data, perform preprocessing, and build a unified evaluation dataset; The hierarchical object modeling module is used to establish four levels of evaluation objects, construct the hierarchical relationships of evaluation objects at each level, and generate a calculation index table. The multi-risk quantification calculation module is used to calculate aging risk, abnormal operation risk, abnormal energy consumption risk, work order deterioration risk, and renovation and upgrading opportunity index respectively. The weighted fusion is used to obtain the health risk index and output the weighted contribution value of each type of risk. The follow-up visit trigger determination module is used to store preset trigger thresholds and iterate through each follow-up visit evaluation object to complete the follow-up visit trigger determination. The work order push module is used to generate follow-up work orders and follow-up notifications, and push the follow-up notifications to the corresponding contacts.

Images