Digital twin operation platform for multi-source data fusion of smart park

Abstract

The application discloses a digital twin operation platform for multi-source data fusion of a smart park, comprising a data acquisition module, a load candidate determination module, a response evaluation module, a deviation determination module and a cause shunting module. The data acquisition module acquires personnel load data, carbon dioxide concentration data and fresh air action data of a target space; the load candidate determination module determines a planned personnel load candidate value and at least one actual personnel load candidate value; the response evaluation module determines a carbon dioxide expected response and a carbon dioxide actual response, respectively; the deviation determination module obtains a response deviation corresponding to each personnel load candidate value; and the cause shunting module determines whether carbon dioxide response deficiency can be explained by actual personnel load exceeding and determines that there is a non-personnel load response anomaly when it cannot be explained. The platform can reduce misjudgments among personnel exceeding, sensor anomalies and ventilation link anomalies, and improve the accuracy of handling fresh air response anomalies in the park.

Description

Technical Field

[0001] This application relates to the field of smart park operation and control technology, and in particular to a digital twin operation platform for the fusion of multi-source data in smart parks. Background Technology

[0002] With the advancement of digital and intelligent construction in industrial parks, smart parks typically deploy multiple subsystems such as meeting reservation systems, access control systems, video recognition systems, building automation systems, carbon dioxide sensor systems, fresh air equipment control systems, and work order systems. They also leverage digital twin operation platforms to centrally display and manage the personnel status, air quality status, equipment operation status, and maintenance status of target spaces such as meeting rooms, training rooms, shared office spaces, laboratories, and negotiation rooms.

[0003] For spaces with fluctuating personnel density, such as conference rooms and training rooms, carbon dioxide concentration can often serve as an important indicator of air quality and ventilation needs. When the carbon dioxide concentration in the target space increases, existing platforms can typically control the opening of the fresh air valve, increase the speed of the fresh air fan, or switch the fresh air operation level based on the carbon dioxide concentration threshold, thereby increasing the fresh air volume and improving the air quality in the target space.

[0004] However, in actual park operations, the personnel load data provided by different systems are often inconsistent. For example, a meeting reservation system might show 10 people booked for a meeting room, while the actual number of people entering could be 20; the access control system might have a lower cumulative number of people due to multiple people following each other in, leaving without swiping their cards, or delayed uploads of access control records; and the video recognition system might identify a number of people on-site that does not match the reserved number or the cumulative number of people accessed. In such cases, when the fresh air system has already activated its enhanced operation but the carbon dioxide concentration does not decrease as expected, existing platforms often struggle to determine whether this insufficient response is due to the actual personnel load exceeding the planned personnel load, or due to a stuck fresh air valve, malfunctioning fresh air unit, clogged filter, malfunctioning duct, or a faulty carbon dioxide sensor.

[0005] While existing technologies can control fresh air volume based on carbon dioxide concentration or adjust the operation of fresh air equipment based on the number of people, they generally rely on a single source of people, a single carbon dioxide threshold, or a single piece of equipment feedback for judgment. When the carbon dioxide response is insufficient, existing platforms tend to directly attribute it to fresh air equipment malfunction or simply indicate that the carbon dioxide concentration exceeds the standard, lacking a mechanism to differentiate the cause based on the correlation between planned personnel load, actual personnel load, fresh air operation, and actual carbon dioxide response. This could lead to overcrowding being misjudged as equipment malfunction, or genuine ventilation link anomalies being misjudged as overcrowding, resulting in ineffective operation of fresh air equipment, persistent abnormal air quality, and incorrect handling of maintenance work orders.

[0006] Therefore, it is necessary to provide a digital twin operation platform for smart parks that integrates multi-source data. This platform can integrate planned personnel load data, actual personnel load data, carbon dioxide concentration data, and fresh air operation data when insufficient carbon dioxide response occurs after the fresh air equipment is activated. It can first determine whether the insufficient response can be explained by the actual personnel load exceeding the limit. Then, for anomalies that cannot be explained by personnel load, it can further carry out diversion work for carbon dioxide sensor anomalies or ventilation link anomalies, so as to improve the accuracy of the smart park operation platform in judging and handling fresh air response anomalies. Summary of the Invention

[0007] This application provides a digital twin operation platform for multi-source data fusion in smart parks. When the fresh air equipment in the target space of the park has performed its operation but the carbon dioxide concentration has not improved as expected, it can distinguish whether the insufficient response is caused by the actual personnel load exceeding the limit or by non-personnel load factors based on the correlation between the planned personnel load, the actual personnel load, the fresh air operation, and the carbon dioxide response. This reduces misjudgment of ventilation anomalies and incorrect handling of the target.

[0008] Firstly, this application provides a digital twin operation platform for multi-source data fusion in smart parks. The platform includes a data acquisition module, a load candidate determination module, a response evaluation module, a deviation determination module, and a cause-based analysis module. The data acquisition module acquires personnel load data, carbon dioxide concentration data, and fresh air operation data for the target space. The personnel load data includes planned personnel load data and actual personnel load data. The fresh air operation data characterizes the actions taken by the fresh air equipment in the target space after the carbon dioxide concentration meets the fresh air triggering condition. The load candidate determination module determines planned personnel load candidate values ​​based on the planned personnel load data and determines at least one actual personnel load candidate value based on the actual personnel load data. The response evaluation module uses the planned personnel load candidate value, at least one actual personnel load candidate value, and fresh air... The system uses action data to determine the corresponding expected carbon dioxide response and the actual carbon dioxide response based on the carbon dioxide concentration data after the fresh air operation. The deviation determination module compares the actual carbon dioxide response with each expected carbon dioxide response to obtain the response deviation corresponding to the planned personnel load candidate value and at least one actual personnel load candidate value. The cause triage module determines whether the insufficient carbon dioxide response can be explained by the actual personnel load exceeding the response condition when the response deviation corresponding to the planned personnel load candidate value meets the insufficient response condition. If it can be explained by the actual personnel load exceeding the response condition, the cause is determined to be the personnel load exceeding the response condition. If it cannot be explained by the actual personnel load exceeding the response condition, it is determined that there is an abnormal response due to non-personnel load.

[0009] In this platform, instead of immediately attributing insufficient carbon dioxide response to a malfunction in the fresh air system, the system first uses a load candidate determination module to generate planned and actual personnel load candidate values. Then, a response assessment module determines the expected carbon dioxide response under different personnel load candidate values. A deviation determination module compares the actual carbon dioxide response with each expected response. Finally, a cause triage module determines whether the insufficient response can be explained by an excessive actual personnel load. This platform allows for different scenarios where the number of appointments, the number of people accessed through access control, and the number of people recognized via video are inconsistent. It first determines whether the insufficient fresh air response is due to an excessive actual personnel load, and then determines whether it is due to a non-personnel load response anomaly. This avoids misjudging overcrowding as equipment malfunction and misjudging abnormal equipment ventilation as overcrowding.

[0010] In one possible design, planned personnel load data includes the number of appointments, while actual personnel load data includes at least one of the cumulative number of people accessed through access control and the number of people identified through video recognition. A load candidate determination module is used to determine the number of appointments as a candidate value for planned personnel load, and to determine at least one of the cumulative number of people accessed through access control and the number of people identified through video recognition as a candidate value for actual personnel load. Through this design, the platform can obtain both the planned usage status and the actual usage status of the target space, ensuring that subsequent determination of the cause of insufficient response no longer relies solely on the number of appointments or the cumulative number of people accessed through access control.

[0011] In one possible design, the load candidate determination module is also used to: determine the planned personnel load candidate value based on the target space's planned open status, seat reservation information, historical baseline numbers for similar time periods, or park management policies when there are no reserved personnel or the number of reserved personnel is null; and determine the cumulative number of people at the access control system based on access control entry records, access control exit records, and the cumulative number of people at the previous statistical time. This design allows the platform to be applicable to target spaces with different reservation management modes, such as meeting rooms, shared office spaces, and training rooms, and reduces traffic disruptions caused by missing reservation data.

[0012] In one possible design, the response assessment module is used to determine the expected decrease in carbon dioxide concentration, the expected rate of decrease, or the expected slowdown in the increase, based on the target space volume, planned personnel load candidate values, at least one actual personnel load candidate value, and fresh air operation data. The fresh air operation data includes at least one of the following: fresh air valve opening, fresh air unit speed, fresh air operating level, fresh air supply volume, and fresh air operation trigger time. With this design, the expected carbon dioxide response is related not only to the number of personnel but also to the target space volume and the operation of the fresh air equipment, thus more accurately reflecting the expected changes in carbon dioxide concentration under the current space and fresh air operation conditions.

[0013] In one possible design, the response evaluation module is also used to: determine the equivalent fresh air supply volume based on the data collected by the air supply volume sensor after the fresh air operation when the target space is equipped with an air supply volume sensor; determine the equivalent fresh air supply volume based on the preset mapping relationship between the fresh air valve opening, fresh air fan speed, fresh air operation level, and air supply volume when the target space is not equipped with an air supply volume sensor; and determine the corresponding expected carbon dioxide response based on the equivalent fresh air supply volume, the volume of the target space, candidate occupant load values, the carbon dioxide concentration when the fresh air operation is triggered, and the outdoor or fresh air inlet carbon dioxide concentration. This design provides an executable data source for determining the expected carbon dioxide response, regardless of whether the target space's air supply volume can be directly collected or not.

[0014] In one possible design, the response assessment module is also used to determine the fresh air activation trigger time based on the fresh air activation data; to determine the starting point of the response observation window based on the fresh air activation trigger time; and to determine the length of the response observation window based on the target space volume and at least one of the fresh air activation data; and to determine the actual carbon dioxide response based on the carbon dioxide concentration data within the response observation window, whereby the actual carbon dioxide response includes at least one of the actual decrease in carbon dioxide, the actual rate of decrease, and the actual rate of deceleration of rise. This design avoids making judgments solely based on instantaneous carbon dioxide values ​​before and after the fresh air activation, instead determining whether air quality has undergone a sustained improvement within the response observation window after the fresh air activation.

[0015] In one possible design, the cause-based routing module is used to: determine that the response deviation corresponding to the planned personnel load candidate value satisfies the inadequate response condition when the actual CO2 response corresponding to the planned personnel load candidate value is lower than the expected CO2 response corresponding to the planned personnel load candidate value, and the response deviation corresponding to the planned personnel load candidate value is greater than the response deviation threshold; determine a credible personnel load candidate value from at least one actual personnel load candidate value, wherein the credible personnel load candidate value is an actual personnel load candidate value that is greater than the planned personnel load candidate value and whose corresponding response deviation satisfies the personnel load explanation condition; when a credible personnel load candidate value exists, determine that the CO2 inadequate response can be explained by the excess of the credible personnel load candidate value relative to the planned personnel load candidate value; when no credible personnel load candidate value exists, determine that the CO2 inadequate response cannot be explained by the excess of the actual personnel load. Through this design, the response deviation corresponding to the actual personnel load candidate value can be used to reversely determine whether the excess of the actual personnel load is sufficient to explain the current CO2 inadequate response.

[0016] In one possible design, the personnel load interpretation criteria include: the response deviation corresponding to the actual personnel load candidate value is less than or equal to the response deviation threshold; or, the response deviation corresponding to the actual personnel load candidate value is less than the response deviation corresponding to the planned personnel load candidate value, and the response deviation corresponding to the actual personnel load candidate value is less than or equal to the response deviation threshold. With this design, when the actual personnel load candidate value is higher than the planned personnel load candidate value and can better explain the actual carbon dioxide response, the cause can be identified as excessive personnel load, rather than directly resulting in a malfunction of the fresh air equipment.

[0017] In one possible design, the cause-based routing module is also used to: after determining that there is an abnormal response to non-human load, acquire at least one of the adjacent carbon dioxide sensor data and historical carbon dioxide change data associated with the target space; when the consistency between the carbon dioxide concentration data of the target space and the adjacent carbon dioxide sensor data or historical carbon dioxide change data does not meet the sensor reliability condition, determine the abnormal response to non-human load as a carbon dioxide sensor abnormality; when the carbon dioxide concentration data of the target space meets the sensor reliability condition, acquire at least one of the fresh air valve feedback status, fresh air unit operating status, filter pressure differential data, duct pressure data, and target space air supply volume feedback data associated with the target space; when the response deviation determined based on the maximum actual human load candidate value is greater than the response deviation threshold, and at least one of the fresh air valve feedback status, fresh air unit operating status, filter pressure differential data, duct pressure data, and target space air supply volume feedback data meets the ventilation link abnormality condition, determine the abnormal response to non-human load as a ventilation link abnormality. This design allows for the investigation of carbon dioxide sensor malfunctions after ruling out reasons related to excessive personnel workload. It also enables the identification of specific objects in the ventilation chain to pinpoint the cause of the malfunction, thereby improving the accuracy of cause triage.

[0018] In one possible design, the platform also includes a twin object mapping module and a response strategy generation module. The twin object mapping module maps personnel overload or non-personnel overload response anomalies to target space objects, fresh air equipment objects, carbon dioxide sensor objects, and their associated links in the digital twin operations platform. The response strategy generation module generates fresh air enhancement control suggestions or space overcrowding alerts when the cause is personnel overload; generates maintenance work orders for fresh air valves, fresh air units, filters, or ducts when non-personnel overload response anomalies are determined to be ventilation link anomalies; and generates carbon dioxide sensor calibration work orders when non-personnel overload response anomalies are determined to be carbon dioxide sensor anomalies. Through this design, the platform can establish a correspondence between cause triage results and twin objects, equipment links, and response strategies, thereby facilitating maintenance personnel to quickly understand the cause of the anomaly and take appropriate measures. Attached Figure Description

[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an undue limitation of this application.

[0020] Figure 1 A schematic diagram of the system architecture of a digital twin operation platform for multi-source data fusion in a smart park, provided in an embodiment of this application; Figure 2 This application provides a schematic diagram of an abnormal fresh air response scenario in a target space. Figure 3 A flowchart of in-platform cause-based traffic routing is provided for an embodiment of this application; Figure 4 A schematic diagram illustrating the calculation of candidate personnel load values ​​and expected carbon dioxide response, provided for an embodiment of this application; Figure 5 This application provides a schematic diagram illustrating the cause of non-human load response anomalies in an embodiment of the present application. Figure 6 This is a schematic diagram illustrating the generation of a digital twin object link tagging and handling strategy, provided in an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings. In the description of the embodiments of this application, terms such as "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance, nor should they be construed as implicitly indicating the number of indicated technical features.

[0022] It should be understood that in the embodiments of this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships, for example, A and / or B, which can represent the cases where A exists alone, A and B exist simultaneously, or B exists alone, where A and B can be singular or plural. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of singular or plural items.

[0023] This application provides a digital twin operation platform for multi-source data fusion in smart parks. (Refer to...) Figure 1 , Figure 1This is a schematic diagram of the system architecture of a digital twin operation platform for multi-source data fusion in a smart park, provided as an embodiment of this application. The platform can communicate with meeting reservation systems, access control systems, video recognition systems, building automation systems, carbon dioxide sensor systems, fresh air equipment control systems, and work order systems to perform fusion processing and digital twin mapping of personnel load status, carbon dioxide concentration status, fresh air equipment operation status, and anomaly handling results in the target space.

[0024] The target space can be a conference room, training room, shared office space, laboratory, meeting room, or other enclosed or semi-enclosed space with fresh air conditioning requirements. Fresh air equipment may include at least one of the following: fresh air valve, fresh air unit, fresh air duct, filter, and airflow regulating device. A carbon dioxide sensor may be installed within the target space, or in a space adjacent to or with airflow access to the target space.

[0025] This smart park's multi-source data fusion digital twin operation platform can include a data acquisition module, a load candidate determination module, a response evaluation module, a deviation determination module, a cause-based routing module, a twin object mapping module, and a disposal strategy generation module. Specifically, the data acquisition module can acquire meeting reservation data, access control data, video recognition data, carbon dioxide concentration data, and fresh air operation data; the load candidate determination module can determine planned and actual personnel load candidate values; the response evaluation module can determine the expected and actual carbon dioxide response; the deviation determination module can determine the response deviation; the cause-based routing module can determine whether personnel load exceeds limits, non-personnel load response is abnormal, carbon dioxide sensor malfunctions, or ventilation link malfunctions; the twin object mapping module can map the cause-based routing results to target space objects, fresh air equipment objects, carbon dioxide sensor objects, and their associated links; and the disposal strategy generation module can generate fresh air enhancement control suggestions, space overcrowding alerts, maintenance work orders, or carbon dioxide sensor calibration work orders.

[0026] Within this platform, target space objects can include space number, space volume, space purpose, associated personnel data sources, associated fresh air equipment objects, and associated carbon dioxide sensor objects. Fresh air equipment objects can include equipment number, equipment type, control commands, feedback status, airflow feedback, and maintenance records. Carbon dioxide sensor objects can include sensor number, installation location, sampling period, and historical concentration data. Using this object information, the platform can, after determining the cause of the anomaly, map the cause to the corresponding target space object, fresh air equipment object, carbon dioxide sensor object, and their associated links.

[0027] In this embodiment, personnel load data can be used to characterize the planned number of users or the actual number of users on-site in the target space. Planned personnel load data may include the number of reservations. Actual personnel load data may include at least one of the following: cumulative number of users accessed through access control and number of users identified via video recognition. Further, actual personnel load data may also include at least one of the following: personnel location data, visitor registration data, seat occupancy sensor data, wireless probe statistics, and turnstile access data. Here, the number of reservations can characterize the planned personnel load of the target space, the cumulative number of users accessed through access control can characterize the personnel load based on entry and exit records, and the number of users identified via video recognition can characterize the on-site personnel load identified by video within the target space.

[0028] In this embodiment of the application, the carbon dioxide concentration data may include a sequence of carbon dioxide concentrations in the target space before and after the fresh air operation. The fresh air operation data can be used to characterize the actions taken by the fresh air equipment after the carbon dioxide concentration meets the fresh air triggering condition. For example, the fresh air operation data may include at least one of the following: fresh air valve opening degree, fresh air fan speed, fresh air operating level, fresh air delivery volume, and fresh air operation trigger time.

[0029] Reference Figure 2 , Figure 2 This is a schematic diagram illustrating an abnormal fresh air response scenario in a target space, as provided in an embodiment of this application. In this scenario, there is a discrepancy between the number of people who made reservations, the number of people recorded through access control, and the number of people identified by video. When the carbon dioxide concentration rises and triggers the fresh air equipment, if the carbon dioxide concentration does not improve as expected under the planned personnel load, the platform needs to determine whether this insufficient response is caused by an actual personnel load exceeding the limit, or by an abnormality in the carbon dioxide sensor or the ventilation chain.

[0030] Reference Figure 3 , Figure 3 This application provides a flowchart for a platform-based cause triage process. The digital twin operation platform for multi-source data fusion in this smart park can use the following processing flow to triage the causes of insufficient fresh air response.

[0031] S101: The data acquisition module acquires personnel load data, carbon dioxide concentration data, and fresh air operation data of the target space; The data acquisition module can obtain the number of registered users for the target space from the meeting reservation system, and the access control system's entry and exit records for the target space. Based on these records, it determines the cumulative number of users at the access control points. For example, the cumulative number of users can be determined based on the cumulative number at the previous statistical time, the number of users entering during the current statistical period, and the number of users leaving during the current statistical period.

[0032] The data acquisition module can also obtain the number of people identified in the video recognition system within the target space. The number of people identified in the video recognition can be the number of people identified by cameras within the target space, or it can be the number of people obtained after deduplicating the recognition results from multiple cameras. This application embodiment does not limit the specific method for identifying the number of people in the video recognition.

[0033] The data acquisition module can obtain carbon dioxide concentration data of the target space from the carbon dioxide sensor system. This carbon dioxide concentration data can include the carbon dioxide concentration before the fresh air operation is triggered, the carbon dioxide concentration after the fresh air operation is triggered, and the carbon dioxide concentration change sequence within a preset time period.

[0034] The data acquisition module can obtain fresh air operation data from the building automation system or the fresh air equipment control system. This fresh air operation data may include at least one of the following: fresh air valve opening degree, fresh air fan speed, fresh air operating level, fresh air supply volume, and fresh air operation trigger time. The fresh air operation trigger time may be the time when the fresh air valve opening degree changes, the time when the fresh air fan speed changes, or the time when the fresh air control command is issued.

[0035] S102: The load candidate determination module determines the planned personnel load candidate value and the actual personnel load candidate value; The load candidate determination module can determine the planned personnel load candidate value based on the number of reservations. For example, if the meeting reservation system shows that the number of reservations for the target space in the current time period is 12, then 12 can be determined as the planned personnel load candidate value.

[0036] When the target space has no reservations or the reservation count is zero, the load candidate determination module can either set the planned personnel load candidate value to zero, or determine the planned personnel load candidate value based on the target space's planned openness status, seat reservation information, historical baseline numbers for similar time periods, or park management policies. Therefore, for shared office spaces, open meeting areas, or training spaces without mandatory reservations, planned personnel load candidate values ​​can also be obtained for subsequent cause-based traffic management.

[0037] The load candidate determination module can determine the actual personnel load candidate value based on at least one of the cumulative number of people accessed through the access control system and the number of people identified through video recognition. For example, if the cumulative number of people accessed through the access control system is 10 and the number of people identified through video recognition is 18, then 10 can be determined as the first actual personnel load candidate value and 18 can be determined as the second actual personnel load candidate value.

[0038] In some implementations, the load candidate determination module can use the cumulative number of people at the access control system and the number of people recognized by video as candidate values ​​for actual personnel load. In other implementations, the load candidate determination module can also determine the fused candidate value for actual personnel load based on the reliability of the cumulative number of people at the access control system and the number of people recognized by video. This application does not limit the number of candidate values ​​for actual personnel load, as long as there is at least one candidate value.

[0039] S103: The load candidate determination module determines whether there are differences in personnel load in the target space; The load candidate determination module can compare the planned personnel load candidate value with any actual personnel load candidate value. When the difference between the planned personnel load candidate value and any actual personnel load candidate value is greater than the personnel difference threshold, the load candidate determination module can determine that there is a personnel load difference in the target space and identify the target space as the target space that needs to be interpreted and judged.

[0040] For example, the personnel difference threshold can be a fixed number threshold, such as 5 people; or it can be a proportional threshold, such as the difference ratio between the actual personnel load candidate value and the planned personnel load candidate value being greater than 30%. If the number of appointments is 12 people, the number of people identified by video is 20 people, and the personnel difference threshold is 5 people, then the load candidate determination module can determine that there is a personnel load difference in the target space.

[0041] In some implementation methods, when there is a difference in the number of people in the target space, it indicates that the number of reservations may not match the actual number of people on site, and the subsequent insufficient carbon dioxide response may have room for interpretation based on the number of people. In this way, misjudgments can be reduced in scenarios where there is no obvious difference in the number of people, and the problem of attributing abnormal fresh air response caused by the discrepancy between the number of reservations and the actual number of people can be addressed.

[0042] S104: The response assessment module determines the expected carbon dioxide response corresponding to each candidate value of personnel load; The response assessment module can determine the expected carbon dioxide response corresponding to the planned personnel load candidate value and the actual personnel load candidate value, respectively, based on the target space volume, the planned personnel load candidate value, at least one actual personnel load candidate value, and fresh air operation data. The expected carbon dioxide response can include the expected decrease in carbon dioxide, the expected rate of decrease, or the expected slowdown in the increase.

[0043] For example, for any candidate personnel load value, the expected trend of carbon dioxide change after the fresh air operation can be determined based on the personnel carbon dioxide generation corresponding to that candidate personnel load value, the spatial volume of the target space, and the fresh air supply volume. Here, personnel carbon dioxide generation can be determined based on the personnel load candidate value and the carbon dioxide generation per unit of personnel; the fresh air supply volume can be determined based on the fresh air valve opening, the fresh air fan speed, or the fresh air operation level.

[0044] Reference Figure 4 , Figure 4 This is a schematic diagram illustrating the calculation of candidate personnel load values ​​and expected carbon dioxide response, provided as an embodiment of this application. In one optional implementation, for the i-th candidate personnel load value Pi, the response evaluation module can determine the expected carbon dioxide decrease rate Rpi corresponding to the i-th candidate personnel load value Pi according to the following formula: Rpi=Qv / V×(Ct0-Cout)-Pi×Gp / V; Where Rpi represents the expected rate of decrease of carbon dioxide corresponding to the i-th candidate value of personnel load, and the unit can be ppm / min; Qv represents the equivalent fresh air supply volume of the fresh air equipment within the response observation window, and the unit can be m³ / min; V represents the spatial volume of the target space, and the unit can be m³; Ct0 represents the carbon dioxide concentration in the target space when the fresh air action is triggered, and the unit can be ppm; Cout represents the carbon dioxide concentration at the outdoor or fresh air inlet, and the unit can be ppm; Pi represents the i-th candidate value of personnel load, and the unit is person; Gp represents the equivalent amount of carbon dioxide concentration generated per person per unit time, and the unit can be ppm·m³ / min / person; Pi×Gp / V represents the rate of increase of carbon dioxide concentration formed by the i-th candidate value of personnel load in the target space.

[0045] When the target space is equipped with an air volume sensor, Qv can be the average fresh air supply volume collected by the air volume sensor within the response observation window. When the target space is not equipped with an air volume sensor, Qv can be determined based on a preset mapping relationship between the fresh air valve opening, fresh air fan speed, fresh air operating level, and air supply volume. This preset mapping relationship can be calibrated using equipment factory parameters, on-site commissioning data, or historical operating data. For example, during the equipment commissioning phase, the platform can record the air supply volume under different combinations of fresh air valve opening and fresh air fan speed, forming a lookup table between action intensity and equivalent fresh air supply volume, and then query the corresponding Qv based on the current fresh air action data during subsequent operation.

[0046] According to the above formula, the stronger the fresh air dilution capacity, the greater the expected rate of carbon dioxide reduction; the greater the personnel load, the greater the corresponding personnel carbon dioxide production term, and the smaller the expected rate of carbon dioxide reduction. When Rpi is positive, it means that the carbon dioxide concentration is expected to decrease under the i-th candidate personnel load value; when Rpi is zero or negative, it means that under the i-th candidate personnel load value, the current fresh air operation is unlikely to reduce the carbon dioxide concentration, and the carbon dioxide concentration may remain unchanged or continue to rise. In this case, Rpi can be converted into the expected rate of decrease or the expected net improvement to participate in the calculation of response deviation.

[0047] In another optional implementation, the response assessment module can call the historical response curve of the target space and determine the expected carbon dioxide response corresponding to the current candidate value of personnel load based on the carbon dioxide change results of the same or similar personnel load, fresh air valve opening, fresh air fan speed and space usage status in history.

[0048] S105: The response assessment module determines the actual carbon dioxide response; The response assessment module can determine the actual carbon dioxide response based on the carbon dioxide concentration data after the fresh air operation. For example, the response assessment module can determine the fresh air operation trigger time based on the fresh air operation data, and determine the starting point of the response observation window based on the fresh air operation trigger time.

[0049] The response assessment module can determine the length of the response observation window based on at least one of the target space's volume and fresh air operation data. For example, the response observation window length can be 5 minutes, 10 minutes, or 15 minutes, or it can be dynamically determined based on the target space's volume, fresh air supply volume, and current carbon dioxide concentration.

[0050] In some implementations, the larger the target space volume or the smaller the fresh air supply volume, the longer the response observation window can be set; conversely, the smaller the target space volume or the larger the fresh air supply volume, the shorter the response observation window can be set. By setting a response observation window, misjudgments can be avoided due to factors such as the fresh air action being triggered immediately, air mixing lag, or different sampling intervals of the carbon dioxide sensor.

[0051] The response assessment module can determine the actual carbon dioxide response based on the carbon dioxide concentration data in the response observation window. The actual carbon dioxide response can include at least one of the following: the actual decrease in carbon dioxide concentration, the actual rate of decrease, and the actual rate of decrease in carbon dioxide concentration.

[0052] For example, if the carbon dioxide concentration in the target space is 1200 ppm when the fresh air action is triggered, and the carbon dioxide concentration is 1120 ppm at the end of the response observation window, then the actual decrease in carbon dioxide concentration could be 80 ppm. If the response observation window is 10 minutes, then the actual decrease rate in carbon dioxide concentration could be 8 ppm / min.

[0053] For example, if the rate of increase in carbon dioxide concentration is 20 ppm / min before the fresh air system is activated, and the rate of increase decreases to 5 ppm / min after the system is activated, then the actual rate of decrease can be 15 ppm / min. For target spaces with high population density where the carbon dioxide concentration is still rising, the actual rate of decrease can be used as the actual carbon dioxide response.

[0054] S106: The deviation determination module calculates the response deviation and determines whether there is insufficient carbon dioxide response; The deviation determination module can compare the actual carbon dioxide response with the expected carbon dioxide response corresponding to each candidate personnel load value, obtaining the response deviations corresponding to the planned candidate personnel load value and at least one actual candidate personnel load value. The response deviation can be used to characterize the degree of difference between the actual air quality change and the expected air quality change under a certain candidate personnel load value.

[0055] In one optional implementation, the deviation determination module can determine the response deviation Di corresponding to the i-th candidate value of personnel load according to the following formula: Di=|Rpi-Ra| / max(|Rpi|,ε; Where Di represents the response deviation corresponding to the i-th candidate value of personnel load; Rpi represents the expected carbon dioxide response corresponding to the i-th candidate value of personnel load, which can be the expected rate of decrease, the expected magnitude of decrease, or the expected rate of decrease in carbon dioxide; Ra represents the actual carbon dioxide response after the fresh air operation, which can be the actual rate of decrease, the actual magnitude of decrease, or the actual rate of decrease in increase in carbon dioxide; ε is a preset response lower limit value to avoid the denominator being zero, and its unit is consistent with Rpi. ε can be preset based on the resolution of the carbon dioxide sensor or the platform calculation accuracy.

[0056] When calculating response deviation, Rpi and Ra use the same type of response metrics. For example, when Rpi is the expected rate of decrease in carbon dioxide, Ra is the actual rate of decrease in carbon dioxide; when Rpi is the expected magnitude of decrease in carbon dioxide, Ra is the actual magnitude of decrease in carbon dioxide; when Rpi is the expected rate of slowdown in the rise of carbon dioxide, Ra is the actual rate of slowdown in the rise of carbon dioxide.

[0057] A smaller response deviation indicates that the corresponding candidate personnel load value can better explain the current actual carbon dioxide response; a larger response deviation indicates that the corresponding candidate personnel load value is less able to explain the current actual carbon dioxide response. When the response deviation is greater than the response deviation threshold, it can be determined that the corresponding candidate personnel load value cannot explain the current actual carbon dioxide response.

[0058] In the embodiments of this application, insufficient carbon dioxide response can refer to a state where, after the fresh air equipment is activated, the actual carbon dioxide response is lower than the expected carbon dioxide response determined based on the candidate value of the planned personnel load, and the response deviation corresponding to the candidate value of the planned personnel load is greater than the response deviation threshold. Insufficient response can manifest as the actual rate of decrease being lower than the expected rate of decrease, the actual magnitude of decrease being lower than the expected magnitude of decrease, or the actual rate of decrease in carbon dioxide rise being lower than the expected rate of decrease in rise.

[0059] In the embodiments of this application, the thresholds corresponding to the personnel difference threshold, response deviation threshold, and sensor reliability condition can be preset values ​​or determined based on historical operating data of the target space. For example, the response deviation threshold can be determined based on the historical response deviation distribution of the target space when the fresh air equipment is operating normally and the personnel load is stable. The threshold can be the historical response deviation mean plus twice the standard deviation, or the 95th percentile of the historical response deviation can be used as the threshold.

[0060] S107: The cause triage module determines whether insufficient carbon dioxide response can be explained by actual personnel load exceeding the limit; The cause-based load decomposition module can determine a credible personnel load candidate value from at least one actual personnel load candidate value. A credible personnel load candidate value is an actual personnel load candidate value that is greater than the planned personnel load candidate value and whose corresponding response deviation meets the personnel load interpretation conditions.

[0061] In some implementations, the personnel load interpretation criteria may include: the response deviation corresponding to the actual personnel load candidate value is less than or equal to a response deviation threshold. In other implementations, the personnel load interpretation criteria may include: the response deviation corresponding to the actual personnel load candidate value is less than the response deviation corresponding to the planned personnel load candidate value, and the response deviation corresponding to the actual personnel load candidate value is less than or equal to a response deviation threshold.

[0062] When a credible personnel load candidate value exists, the cause triage module can determine that the insufficient carbon dioxide response can be explained by the excess of the credible personnel load candidate value relative to the planned personnel load candidate value, and identify the cause as excessive personnel load. In this case, the insufficient carbon dioxide response is not because the fresh air equipment itself is not working, but because the actual number of people on site is higher than the planned number, resulting in the current fresh air operation intensity being insufficient to meet the actual personnel load on site.

[0063] When no credible personnel load candidate value exists, the cause triage module can determine that insufficient carbon dioxide response cannot be explained by an over-expansion of actual personnel load and identify a non-personnel load response anomaly. A non-personnel load response anomaly indicates that insufficient carbon dioxide response cannot be explained by an over-expansion of actual personnel load relative to planned personnel load. Non-personnel load response anomalies may further include carbon dioxide sensor malfunctions or ventilation link malfunctions.

[0064] S108: The cause is that the shunt module determined that the carbon dioxide sensor was malfunctioning; Reference Figure 5 , Figure 5 This is a schematic diagram illustrating the cause-based routing for non-personnel load response anomalies provided in this application embodiment. In some implementations, the cause-based routing module can perform a carbon dioxide sensor reliability check after determining that insufficient carbon dioxide response cannot be explained by actual personnel load exceeding the limit. When the carbon dioxide concentration data does not meet the sensor reliability condition, the non-personnel load response anomaly is identified as a carbon dioxide sensor anomaly; when the carbon dioxide concentration data meets the sensor reliability condition, the system then uses the fresh air valve feedback status, fresh air unit operating status, filter differential pressure data, duct pressure data, and target space airflow feedback data to determine whether there is a ventilation link anomaly.

[0065] The cause-based triage module can acquire at least one of the following: neighboring carbon dioxide sensor data and historical carbon dioxide change data associated with the target space. When the consistency between the carbon dioxide concentration data of the target space and the neighboring carbon dioxide sensor data or the historical carbon dioxide change data does not meet the sensor reliability condition, the non-human load response anomaly can be identified as a carbon dioxide sensor anomaly.

[0066] The sensor reliability conditions may include at least one of the following: the direction of change of carbon dioxide concentration in the target space is consistent with the direction of change of nearby carbon dioxide sensors; the difference between the change in carbon dioxide concentration in the target space and the change in the same historical load does not exceed the reliability difference threshold; after the fresh air operation intensity changes, the carbon dioxide concentration data in the target space does not remain unchanged for a long time; the sudden change in carbon dioxide concentration in the target space does not exceed the sensor sudden change threshold.

[0067] For example, if the carbon dioxide sensor in the target space maintains a fixed reading for an extended period, while data from neighboring carbon dioxide sensors, video-recognized personnel data, and fresh air operation data all indicate that the carbon dioxide concentration in the target space should have changed significantly, then it can be determined that the carbon dioxide sensor is malfunctioning. As another example, if the carbon dioxide concentration in the target space suddenly increases, but data from neighboring sensors, personnel data, and historical data do not support this upward trend, then it can be determined that the carbon dioxide sensor is drifting or experiencing false alarms.

[0068] S109: The cause of the ventilation link is abnormal, as determined by the diversion module. When the carbon dioxide concentration data of the target space meets the sensor's reliability conditions, the cause diversion module can obtain at least one of the following: the feedback status of the fresh air valve associated with the target space, the operating status of the fresh air unit, the filter differential pressure data, the duct pressure data, and the air supply volume feedback data of the target space, and determine whether there is an abnormality in the ventilation link based on the above data.

[0069] In some implementations, when the response deviation determined based on the maximum actual personnel load candidate value is greater than the response deviation threshold, and at least one of the following—fresh air valve feedback status, fresh air unit operating status, filter differential pressure data, duct pressure data, and target space air supply volume feedback data—meets the ventilation link abnormality condition, the cause diversion module can determine the non-personnel load response abnormality as a ventilation link abnormality.

[0070] For example, if the platform issues an opening command to the fresh air valve, and the valve's opening degree is less than the target opening degree, or the valve's opening degree remains unchanged for an extended period, the fresh air valve is considered abnormal. If the fresh air unit's speed feedback value is lower than the target speed, or the fresh air unit's status display shows a fault, the fresh air unit is considered abnormal. If the filter pressure difference exceeds the filter clogging threshold, or if multiple target spaces experience insufficient fresh air response when both the fresh air valve and fresh air unit are functioning normally, the filter clogging is considered abnormal. If the fresh air unit is operating normally and the fresh air valve feedback is normal, but the target space's airflow feedback is lower than the target airflow, or adjacent spaces respond normally while the target space responds abnormally, the duct is considered abnormal.

[0071] In the embodiments of this application, the filter pressure difference data can be the pressure difference across the filter or the pressure difference detection data of the fresh air unit where the filter is located; the duct pressure data can be the pressure data of the fresh air branch pipe or main pipe corresponding to the target space; the target space air volume feedback data can be the air volume sensor data at the air outlet of the target space or the air volume feedback data estimated based on the valve opening, fan speed and historical calibration relationship.

[0072] S110: The twin object mapping module and the disposal strategy generation module perform object mapping and disposal strategy generation; Reference Figure 6 , Figure 6 This illustration shows a digital twin object link tagging and handling strategy generation method provided in this application embodiment. The twin object mapping module can map personnel overload or non-personnel overload response anomalies to target space objects, fresh air equipment objects, carbon dioxide sensor objects, and their associated links in the platform. For example, the target space object can be a conference room object, the fresh air equipment object can include a fresh air valve object, a fresh air unit object, a filter object, or a duct object, and the carbon dioxide sensor object can be a carbon dioxide sensor installed in the target space.

[0073] When the cause is excessive personnel load, the response strategy generation module can generate suggestions for enhanced fresh air control or space overcrowding alerts. For example, the platform can indicate that the actual number of people in the target space exceeds the reserved number and suggest increasing the fresh air operation level, adjusting meeting room usage arrangements, or alerting management personnel to space overcrowding.

[0074] When a non-personnel load response anomaly is determined to be a ventilation link anomaly, the handling strategy generation module can generate maintenance work orders for fresh air valves, fresh air units, filters, or ducts. For example, if the anomaly is determined to be a fresh air valve, the work order object can be the fresh air valve associated with the target space; if the anomaly is determined to be a filter blockage, the work order object can be the filter of the corresponding fresh air unit; if the anomaly is determined to be a duct, the work order object can be the fresh air duct corresponding to the target space.

[0075] When a non-personnel load response anomaly is determined to be a carbon dioxide sensor malfunction, the handling strategy generation module can generate a carbon dioxide sensor calibration work order. This work order can include the sensor location, anomaly time, anomaly reading, comparison results with nearby sensors, and suggested calibration methods.

[0076] To facilitate understanding, the following are various specific application examples: A conference room has a volume of 120 m³, with 10 people scheduled for appointments, 11 people logged in via access control, and 22 people logged in via video recognition. When the carbon dioxide concentration reaches 1200 ppm, the fresh air valve opening increases from 30% to 80%, the fresh air fan speed increases from low to high, the response evaluation module determines the equivalent fresh air supply volume to be 6 m³ / min, and sets the response observation window to 10 minutes.

[0077] The response assessment module calculated the expected carbon dioxide reduction rate for 10, 11, and 22 people respectively, obtaining expected reduction rates of 15 ppm / min, 14 ppm / min, and 5 ppm / min. The actual reduction rate within the response observation window was 4.8 ppm / min. If the response deviation threshold is 0.2, the deviation determination module determines the response deviation for 10 people to be |15-4.8| / 15=0.68, for 11 people to be |14-4.8| / 14=0.66, and for 22 people to be |5-4.8| / 5=0.04. Since the response deviation for 22 people is less than the response deviation threshold, and 22 people is greater than the number of people scheduled (10), the cause triage module determines that the insufficient carbon dioxide response can be explained by the actual personnel load exceeding the limit. The cause is determined to be personnel load exceeding the limit, and the handling strategy generation module generates a space overcrowding warning and a fresh air enhancement control suggestion.

[0078] In another scenario, a meeting room had 12 people booked, 12 people logged in via access control, and 13 people logged in via video recognition. When the carbon dioxide concentration reached 1250 ppm, the platform issued a control command to the fresh air valve to increase its opening from 30% to 80% and switched the fresh air unit to a high speed. The response evaluation module calculated the expected rate of decrease in carbon dioxide at 12 ppm / min based on the maximum actual personnel load candidate value of 13 people. However, the actual rate of decrease within the response observation window was only 1 ppm / min, corresponding to a response deviation greater than the response deviation threshold.

[0079] The cause-based triage module further retrieved the feedback status of the fresh air valve and found that the valve's opening remained at 35% for an extended period, failing to reach the target opening of 80%. Simultaneously, the carbon dioxide sensor data met the sensor's reliability criteria. At this point, the cause-based triage module determined that the insufficient carbon dioxide response could not be explained by actual personnel load exceeding limits, and identified the abnormal response due to non-personnel load as a fresh air valve malfunction. The handling strategy generation module then generated a maintenance work order for the specific fresh air valve.

[0080] In another scenario, a training room had 20 people booked, 21 people logged in through access control, and 20 people logged in through video recognition. After the carbon dioxide concentration reached 1300 ppm, the fresh air valve feedback indicated that the opening had reached the target opening, and the fresh air unit speed feedback value also reached the high speed. However, the actual rate of decrease in carbon dioxide within the response observation window was still significantly lower than the expected rate of decrease calculated based on the candidate value of the maximum actual personnel load.

[0081] The cause-based triage module obtains filter pressure difference data and finds that the pressure difference across the filter exceeds the filter clogging threshold, and multiple target spaces served by the same fresh air unit show insufficient carbon dioxide reduction. Since the carbon dioxide sensor data meets the sensor reliability criteria, and the fresh air valve and fresh air unit provide normal feedback, the cause-based triage module determines the abnormal response of non-personnel load as an abnormal filter clogging. The handling strategy generation module generates a maintenance work order for the corresponding fresh air unit filter.

[0082] In another scenario, a shared office space had no available reservations. The load candidate determination module determined the planned personnel load candidate value based on the planned opening status of the shared office space and the baseline number of people during similar historical periods. The cumulative number of people accessed was 8, and the number of people identified via video was 9. The response evaluation module calculated the expected rate of decrease in carbon dioxide based on the maximum actual personnel load candidate value of 9 people, which was 9 ppm / min. However, the actual rate of decrease within the response observation window was only 2 ppm / min. The cause diversion module further confirmed that the carbon dioxide sensor data met the sensor reliability conditions, the fresh air valve feedback status was normal, the fresh air unit was operating normally, and the filter pressure difference data did not exceed the filter clogging threshold.

[0083] The cause-based flow analysis module obtains feedback data on the air supply volume of the target space and finds that the actual air supply volume at the air outlet of the shared office space is lower than the target air supply volume. At the same time, the carbon dioxide concentration in adjacent spaces decreases normally under the same fresh air unit operation conditions. Based on this, the cause-based flow analysis module determines that the fresh air branch pipe corresponding to the target space may have duct blockage, air leakage, or abnormal air volume distribution. It identifies the abnormal response to non-personnel load as a duct abnormality, and the handling strategy generation module generates a maintenance work order for the fresh air branch pipe of the target space.

[0084] In another scenario, the carbon dioxide sensor reading in the target space remained constant at 1200 ppm for an extended period before and after the fresh air intake operation, while data from nearby sensors and historical carbon dioxide change data indicated that the carbon dioxide concentration should have decreased under the same fresh air intake intensity. Simultaneously, the number of people identified through video recognition and the cumulative number of people accessed through access control did not show a significant increase. The cause triage module determined that the carbon dioxide concentration data in the target space did not meet the sensor's reliability criteria, identifying the abnormal non-personnel load response as a carbon dioxide sensor malfunction. The handling strategy generation module then generated a carbon dioxide sensor calibration work order.

[0085] In another scenario, a meeting room has 8 people booked, 7 people logged in through access control, and 16 people logged in through video recognition. The response evaluation module calculates an expected carbon dioxide reduction rate of 16 ppm / min based on the 8 booked people, 17 ppm / min based on the 7 logged-in people, and 7 ppm / min based on the 16 logged-in people. The actual reduction rate observed in the response observation window is 6.5 ppm / min.

[0086] The deviation determination module calculated that the response deviations for both the number of appointments and the cumulative number of access control users were greater than the response deviation threshold, while the response deviation for the number of users identified through video recognition was less than the threshold. Based on this, the cause-based triage module identified the number of users identified through video recognition as a reliable candidate value for personnel load, determined the cause as excessive personnel load, and marked a low cumulative number of access control users in the digital twin object. This notification can be used to remind managers to pay attention to situations such as multiple people following others into the room, leaving without swiping their cards, or delayed data uploads from access control, but it does not affect the fact that the cause-based triage result is determined to be excessive personnel load.

[0087] In another scenario, a meeting room has 4 scheduled attendees, 4 attendees according to access control, and 4 attendees identified by video recognition. After the carbon dioxide concentration reaches 1000 ppm, the fresh air valve opening is increased from 20% to 50%. The response assessment module calculates an expected carbon dioxide decrease rate of 6 ppm / min corresponding to the planned personnel load candidate value, while the actual decrease rate within the response observation window is 5.8 ppm / min. The deviation determination module calculates that the response deviation corresponding to the planned personnel load candidate value is less than the response deviation threshold.

[0088] At this point, the cause triage module determines that there is no insufficient carbon dioxide response after the fresh air equipment operates, and the handling strategy generation module does not generate a space overcrowding warning, ventilation link maintenance work order, or carbon dioxide sensor calibration work order. This example demonstrates that the embodiment of this application does not trigger abnormal cause triage in all scenarios of increased carbon dioxide concentration. Instead, it further determines whether the insufficient response can be explained by the actual personnel load exceeding the limit only when the response deviation corresponding to the planned personnel load candidate value meets the insufficient response condition.

[0089] It is understood that the numerical values ​​in the specific examples above are only used to illustrate the implementation of the embodiments of this application and do not constitute a limitation on the scope of protection of this application. In practical applications, space volume, personnel difference threshold, response deviation threshold, response observation window, fresh air supply volume, equivalent amount of carbon dioxide concentration generated per person per unit time, filter clogging threshold, duct pressure threshold, air supply volume feedback threshold, and sensor reliability conditions can all be configured or calibrated based on the target space type, historical operating data, equipment capabilities, and park management strategies.

[0090] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of protection of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A digital twin operation platform for multi-source data fusion in a smart park, characterized in that, include: The data acquisition module is used to acquire personnel load data, carbon dioxide concentration data, and fresh air operation data of the target space. The personnel load data includes planned personnel load data and actual personnel load data. The fresh air operation data is used to characterize the operation of the fresh air equipment in the target space after the carbon dioxide concentration meets the fresh air triggering condition. The load candidate determination module is used to determine the planned personnel load candidate value based on the planned personnel load data, and to determine at least one actual personnel load candidate value based on the actual personnel load data; The response assessment module is used to determine the corresponding expected carbon dioxide response based on the planned personnel load candidate value, the at least one actual personnel load candidate value, and the fresh air operation data, and to determine the actual carbon dioxide response based on the carbon dioxide concentration data after the fresh air operation. The deviation determination module is used to compare the actual carbon dioxide response with each of the expected carbon dioxide responses to obtain the response deviations corresponding to the planned personnel load candidate value and the at least one actual personnel load candidate value, respectively. The cause triage module is used to determine whether the carbon dioxide response insufficiency can be explained by the actual personnel load exceeding the response deviation based on the response deviation corresponding to the at least one actual personnel load candidate value when the response deviation corresponding to the planned personnel load candidate value meets the response insufficiency condition. If the response insufficiency can be explained by the actual personnel load exceeding the response deviation, the cause is determined to be personnel load exceeding the response deviation. If the response insufficiency cannot be explained by the actual personnel load exceeding the response deviation, the cause is determined to be non-personnel load response anomaly.

2. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 1, characterized in that, The planned personnel load data includes the number of appointments, and the actual personnel load data includes at least one of the cumulative number of people accessed through access control and the number of people identified through video recognition. The load candidate determination module is used to determine the number of reservations as the planned personnel load candidate value, and to determine at least one of the cumulative number of people at the access control and the number of people identified by video as the actual personnel load candidate value.

3. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 2, characterized in that, The load candidate determination module is also used for: When the target space has no reservations or the number of reservations is empty, the candidate value of the planned personnel load is determined based on the planned opening status of the target space, seat reservation information, the benchmark number of people in the same historical time period, or the park management strategy. The cumulative number of people at the access control system is determined based on the access control entry records, access control exit records, and the cumulative number of people at the previous statistical time.

4. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 3, characterized in that, The response evaluation module is used for: Based on the spatial volume of the target space, the planned personnel load candidate value, the at least one actual personnel load candidate value, and the fresh air operation data, determine the expected decrease in carbon dioxide, the expected rate of decrease, or the expected slowdown in the rise corresponding to the planned personnel load candidate value and the at least one actual personnel load candidate value, respectively. The fresh air operation data includes at least one of the following: fresh air valve opening, fresh air fan speed, fresh air operation level, fresh air supply volume, and fresh air operation trigger time.

5. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 4, characterized in that, The response evaluation module is also used for: When the target space is equipped with an air supply volume sensor, the equivalent fresh air supply volume is determined based on the data collected by the air supply volume sensor after the fresh air operation. When no air volume sensor is installed in the target space, the equivalent fresh air supply volume is determined according to the preset mapping relationship between the fresh air valve opening, the fresh air fan speed, the fresh air operating level and the air volume. Based on the equivalent fresh air supply volume, the spatial volume of the target space, the planned personnel load candidate value and the at least one actual personnel load candidate value, the carbon dioxide concentration when the fresh air action is triggered, and the outdoor or fresh air inlet carbon dioxide concentration, the corresponding expected carbon dioxide response is determined.

6. The digital twin operation platform for multi-source data fusion in smart parks as described in any one of claims 1 to 5, characterized in that, The response evaluation module is also used for: The fresh air action trigger time is determined based on the fresh air action data. The starting point of the response observation window is determined based on the fresh air action trigger time, and the length of the response observation window is determined based on at least one of the spatial volume of the target space and the fresh air action data. Based on the carbon dioxide concentration data within the response observation window, the actual carbon dioxide response is determined, and the actual carbon dioxide response includes at least one of the actual decrease in carbon dioxide, the actual rate of decrease, and the actual rate of slowdown in rise.

7. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 1, characterized in that, The cause-based traffic splitting module is used for: When the actual carbon dioxide response corresponding to the candidate value of the planned personnel load is lower than the expected carbon dioxide response corresponding to the candidate value of the planned personnel load, and the response deviation corresponding to the candidate value of the planned personnel load is greater than the response deviation threshold, it is determined that the response deviation corresponding to the candidate value of the planned personnel load meets the insufficient response condition. A reliable personnel load candidate value is determined from the at least one actual personnel load candidate value, wherein the reliable personnel load candidate value is an actual personnel load candidate value that is greater than the planned personnel load candidate value and whose corresponding response deviation satisfies the personnel load interpretation condition; When the credible personnel load candidate value exists, it is determined that the insufficient carbon dioxide response can be explained by the excess of the credible personnel load candidate value relative to the planned personnel load candidate value; When no credible personnel load candidate value exists, it is determined that the insufficient carbon dioxide response cannot be explained by the actual personnel load exceeding the limit.

8. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 7, characterized in that, The personnel load interpretation conditions include: The response deviation corresponding to the actual personnel load candidate value is less than or equal to the response deviation threshold; Alternatively, the response deviation corresponding to the actual personnel load candidate value is less than the response deviation corresponding to the planned personnel load candidate value, and the response deviation corresponding to the actual personnel load candidate value is less than or equal to the response deviation threshold.

9. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 8, characterized in that, The cause-based traffic splitting module is also used for: After determining that the non-human load response anomaly exists, acquire at least one of the nearby carbon dioxide sensor data and historical carbon dioxide change data associated with the target space; When the consistency between the carbon dioxide concentration data of the target space and the data of the neighboring carbon dioxide sensor or the historical carbon dioxide change data does not meet the sensor reliability condition, the abnormal non-personnel load response is determined to be a carbon dioxide sensor abnormality. When the carbon dioxide concentration data of the target space meets the sensor's reliability condition, acquire at least one of the following: the fresh air valve feedback status, the fresh air unit operating status, the filter differential pressure data, the duct pressure data, and the target space air supply volume feedback data associated with the target space. When the response deviation determined based on the maximum actual personnel load candidate value is greater than the response deviation threshold, and at least one of the fresh air valve feedback status, fresh air unit operating status, filter pressure difference data, duct pressure data, and target space air supply volume feedback data meets the ventilation link abnormality conditions, the non-personnel load response abnormality is determined as a ventilation link abnormality.

10. The digital twin operation platform for multi-source data fusion in smart parks as described in claim 9, characterized in that, It also includes a twin object mapping module and a disposal strategy generation module; The twin object mapping module is used to map the excessive personnel load or the abnormal response of the non-personnel load to the target space object, fresh air equipment object, carbon dioxide sensor object and their associated links in the digital twin operation platform; The handling strategy generation module is used to generate fresh air enhancement control suggestions or space overcrowding prompts when the cause is excessive personnel load; to generate maintenance work orders for fresh air valves, fresh air units, filters, or ducts when the non-personnel load response anomaly is determined to be a ventilation link anomaly; and to generate a carbon dioxide sensor calibration work order when the non-personnel load response anomaly is determined to be a carbon dioxide sensor anomaly.

Images