Multi-system cooperative control method and system applied to building intelligence

Abstract

The application provides a multi-system cooperative control method and system applied to building intellectualization, relates to the technical field of building intellectualization, and first collects target state feature sets of parameter streams of running states of each subsystem in a building, contains running modes, resource occupation and interactive response features, matches and analyzes the target state feature sets with preset cooperative control constraint conditions, generates a cooperative correlation coefficient set among the subsystems, constructs a cooperative regulation target model based on the cooperative correlation coefficient set among the subsystems and the target state feature sets, and the variables of the cooperative regulation target model contain running parameter regulation amounts and regulation time sequence parameters. An optimal regulation parameter combination is obtained by solving the cooperative regulation target model, is converted into a control instruction sequence, is sent to a subsystem execution end, and running state feedback information is used for updating constraint conditions, so that efficient cooperative control of building multi-systems is realized, and the intellectualization level is improved.

Description

Technical Field

[0001] This invention relates to the field of building intelligence technology, and more specifically, to a multi-system collaborative control method and system applied to building intelligence. Background Technology

[0002] In the field of building intelligence, with the continuous integration of various advanced technologies, buildings typically deploy multiple subsystems, such as lighting systems, air conditioning systems, and security systems, to achieve automated management and efficient operation. However, currently, most of these subsystems operate independently or only perform simple linkage control, lacking a comprehensive and in-depth collaborative working mechanism.

[0003] Traditional building subsystem control methods primarily rely on the operating rules and preset parameters of each subsystem, making efficient information sharing and interaction difficult between them. For example, lighting systems are controlled solely based on light intensity and time, while air conditioning systems adjust cooling or heating power only according to indoor temperature, without considering the impact of each other's operating status. This independent operating mode leads to low overall building energy efficiency and potential functional conflicts or resource waste between different subsystems. For instance, in densely populated areas, lighting and air conditioning systems may operate at high loads simultaneously without dynamic adjustments based on actual occupant distribution and activity, resulting in unnecessary energy consumption. Furthermore, the lack of a unified collaborative control model prevents timely and effective coordinated responses from subsystems when changes occur in the building's operating environment or user needs, hindering the achievement of the comprehensive requirements of building intelligence for efficiency, comfort, and energy conservation. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a multi-system collaborative control method applied to building intelligence, the method comprising: Collect the operational status parameter streams of each subsystem within the building and generate a target status feature set, which includes the operational mode features, resource occupancy features, and interactive response features of each subsystem. The target state feature set is matched and analyzed with the preset collaborative control constraints to generate a collaborative correlation coefficient set between subsystems. The collaborative correlation coefficient set is used to characterize the degree of mutual influence between different subsystems during operation. A collaborative adjustment target model is constructed based on the collaborative correlation coefficient set and the target state feature set. The variables of the collaborative adjustment target model include the adjustment amount of the operating parameters and the adjustment time sequence parameters of each subsystem. Solving the collaborative adjustment target model yields the optimal combination of adjustment parameters that satisfies the constraints. The optimal combination of adjustment parameters includes the specific adjustment values ​​and execution order of each subsystem. The optimal combination of adjustment parameters is converted into a sequence of control instructions that can be executed by the subsystem and sent to the corresponding subsystem execution terminal. The operating status feedback information generated after the control instruction sequence is executed is used to update the preset cooperative control constraints.

[0005] In another aspect, embodiments of the present invention also provide a multi-system collaborative control system for building intelligence, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code, and the processor is used to execute the programs, instructions, or code in the machine-readable storage medium to implement the above-described method.

[0006] Based on the above, this embodiment of the invention generates a target state feature set by collecting the operating state parameter streams of each subsystem within the building. The target state feature set is then matched and analyzed with preset collaborative control constraints to generate a collaborative correlation coefficient set. This accurately characterizes the degree of mutual influence between different subsystems during operation, quantifying the correlation between subsystems. A collaborative adjustment target model is constructed based on the collaborative correlation coefficient set and the target state feature set, clarifying variables such as the adjustment amount and timing parameters of each subsystem's operating parameters. Solving this collaborative adjustment target model yields the optimal combination of adjustment parameters that satisfies the constraints, ensuring that the adjustment scheme meets the operating requirements of each subsystem while achieving optimal operation of the building as a whole. The optimal combination of adjustment parameters is transformed into an executable control command sequence and sent to the corresponding subsystem execution end, realizing efficient collaborative control of multiple systems. Simultaneously, the operating state feedback information is used to update the preset collaborative control constraints, forming a closed-loop control mechanism. This enables the system to continuously adapt to changes in the building's operating environment, continuously improving the building's intelligence level and achieving efficient, comfortable, and energy-saving building operation goals. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the execution flow of the multi-system collaborative control method for building intelligence provided in the embodiments of the present invention.

[0008] Figure 2 This is a schematic diagram of exemplary hardware and software components of a multi-system collaborative control system for building intelligence provided in an embodiment of the present invention. Detailed Implementation

[0009] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a multi-system collaborative control method for building intelligence provided in one embodiment of the present invention. The following is a detailed description of this multi-system collaborative control method for building intelligence.

[0010] Example Implementation Section: Step S110: Collect the operating status parameter stream of each subsystem in the building and generate a target status feature set, which includes the operating mode features, resource occupancy features and interactive response features of each subsystem.

[0011] In this embodiment, multiple subsystems within a commercial office building serve as the application scenario. These subsystems include an air conditioning system, a lighting system, an elevator system, and a security system. The collection of operational status parameter streams for each subsystem is achieved through sensors and data acquisition modules deployed on each subsystem. For the air conditioning system, the operational status parameter stream includes return air temperature, outlet air temperature, air volume, compressor operating frequency, and fan speed; for the lighting system, it includes the on / off status, brightness, and power consumption of lights in each area; for the elevator system, it includes the elevator's floor, direction of travel, number of passengers in the car, travel speed, and door opening / closing time; and for the security system, it includes the working status of each surveillance camera, the on / off status of the access control system, and alarm signal triggering status.

[0012] During the data acquisition process, the aforementioned operational status parameter streams need to be acquired in real time. The acquisition frequency is set according to the characteristics of the subsystem to ensure that changes in the subsystem's operational status can be reflected in a timely manner. Simultaneously, the acquired parameter streams require preliminary preprocessing, such as removing outliers and using interpolation to supplement missing data, in order to ensure the integrity and continuity of the parameter streams.

[0013] After the parameter stream is collected, a target state feature set is generated, which is a characteristic description of the operating state of each subsystem.

[0014] Step S111: Perform feature parsing on the collected operating status parameter streams of each subsystem, and extract pattern recognition feature values ​​from the parameter streams. The pattern recognition feature values ​​include parameter fluctuation period, parameter change trend, and parameter correlation.

[0015] For air conditioning systems, analyze their operating status parameter stream. Parameter fluctuation period refers to the time interval within which parameters such as return air temperature and compressor operating frequency exhibit periodic changes. By analyzing parameter data over a continuous period, the periodic pattern of these fluctuations can be determined. Parameter change trend refers to whether these parameters show an increasing, decreasing, or stable trend over time. For example, during peak office hours, due to increased indoor occupancy, return air temperature may show an upward trend. Parameter correlation refers to the degree of interrelation between different parameters, such as the correlation between changes in supply air volume and changes in return air temperature.

[0016] For lighting systems, parameter fluctuation period can refer to the periodicity of lamp brightness changes over time; for example, brightness may vary regularly at different times of the week. Parameter change trend can refer to the adjustment trend of indoor lighting brightness as outdoor light intensity changes. Parameter correlation can refer to the correlation between changes in lamp brightness in different areas, such as whether there are coordinated changes in lamp brightness in adjacent offices.

[0017] The parameter fluctuation period of an elevator system can refer to the cycle of the elevator's round trip within different time periods, such as the morning rush hour and evening rush hour, during which the elevator's cycle may differ. The parameter change trend can refer to the trend of elevator speed change over time; for example, during off-peak hours, the elevator may run at a lower speed. The parameter correlation can refer to the degree of correlation between the floor the elevator is on and the number of people in the car.

[0018] While the fluctuation cycle of security system parameters may not be obvious, it's possible to analyze the switching cycle of surveillance camera operating states and the changing cycle of access control system open / closed states. Parameter change trends can refer to the changing trends of the security system's alert level over different time periods. Parameter correlation can refer to the degree of correlation between the access control system's open / closed state and alarm signal triggering.

[0019] Step S112: Generate operating mode features based on the pattern recognition feature values. The operating mode features include continuous operating mode identifiers, intermittent operating mode identifiers, and alternating operating mode identifiers. Each mode identifier corresponds to a set of pattern recognition feature value ranges.

[0020] Taking an air conditioning system as an example, its operating mode characteristics are determined based on the extracted pattern recognition feature values. If the air conditioning system's parameter fluctuation cycle is stable, the parameter change trend is relatively gentle, and the parameter correlation is high, conforming to the pattern recognition feature value range corresponding to continuous operation mode, then a continuous operation mode identifier is added to it. This situation typically occurs during office hours, when the air conditioner needs to run continuously to maintain a comfortable indoor temperature.

[0021] In areas where lighting systems are supplemented by natural light during the day, they may operate intermittently. Their parameter fluctuations are not fixed, and the trend of parameter changes shows clear start-stop variations. The parameter correlation is relatively low, conforming to the characteristic value range of intermittent operation mode; therefore, an intermittent operation mode indicator is added.

[0022] Elevator systems may exhibit alternating operating modes at different times. For example, during peak commuting hours, elevators operate continuously at high frequency and speed, while during off-peak hours, they may enter a low-frequency intermittent operating state. The parameter fluctuation cycle, trend, and correlation alternate between these two states, which conforms to the characteristic value range of alternating operating modes. Therefore, an alternating operating mode identifier is added.

[0023] Step S1121: Set the division interval of the pattern recognition feature value. The division interval is determined based on the statistical results of the historical operation data of the building subsystem and includes continuous mode interval, intermittent mode interval and alternating mode interval.

[0024] For air conditioning systems, historical operating data from the past year is analyzed to statistically determine the distribution of fluctuation periods, trends, and correlations of parameters such as return air temperature and compressor operating frequency. In the continuous mode range, the parameter fluctuation period is set to a narrow range; for example, the fluctuation period remains stable within a certain time frame, the slope of the trend is within a small range, and the parameter correlation is within a high range. In the intermittent mode range, the parameter fluctuation period is irregular, the trend shows obvious abrupt changes, and the parameter correlation is within a low range. The alternating mode range is a transitional zone between continuous and intermittent modes, where parameter characteristics alternate between those of continuous and intermittent modes.

[0025] Similarly, for lighting systems, elevator systems, and security systems, corresponding continuous mode intervals, intermittent mode intervals, and alternating mode intervals are set based on their respective historical operation data statistics.

[0026] Step S1122: Compare the extracted pattern recognition feature values ​​with the division interval. When the parameter fluctuation period is in the continuous mode interval, the parameter change trend is linear, and the parameter correlation is in the corresponding division interval, it is marked as a continuous operation mode identifier.

[0027] Taking the operation of the air conditioning system during normal office hours as an example, the extracted parameter fluctuation cycle falls within the continuous mode range, and the change trend of parameters such as return air temperature shows a linear change. For example, as the outdoor temperature rises slowly, the return air temperature also rises linearly. At the same time, the correlation between the air supply volume and the return air temperature is within the range corresponding to the continuous mode. At this time, the air conditioning system is marked as a continuous operation mode.

[0028] Step S1123: When the parameter fluctuation cycle is in the intermittent mode range, the parameter change trend is step-like, and the parameter correlation is in the corresponding division range, it is marked as an intermittent operation mode identifier.

[0029] In some conference room areas, the lighting system may be turned on intermittently according to the meeting schedule. Its parameter fluctuation cycle is within the intermittent mode range, and the brightness value changes in a step-like trend, that is, it suddenly increases during the meeting and suddenly decreases after the meeting ends. The parameter correlation between different lamps is within the range corresponding to the intermittent mode. This is marked as an intermittent operation mode.

[0030] Step S1124: When the parameter fluctuation cycle alternates between the continuous mode interval and the intermittent mode interval, the parameter change trend switches periodically, and the parameter correlation degree changes alternately with the cycle, it is marked as an alternating operation mode identifier.

[0031] The elevator system operates according to the above characteristics on weekdays. During the morning and evening peak hours, the parameter fluctuation cycle is in the continuous mode range, the operating trend is stable, and the parameter correlation is high. However, during the lunch break and nighttime hours, the parameter fluctuation cycle is in the intermittent mode range, the operating trend has obvious pauses, and the parameter correlation is low. The parameter fluctuation cycle alternates between these two ranges, and the changing trend also switches periodically. The parameter correlation also changes periodically with the cycle, therefore it is marked as an alternating operation mode indicator.

[0032] Step S1125: Bind the corresponding pattern recognition feature value range to each operating mode identifier to form an operating mode feature that includes the identifier type and feature range. The feature range includes the interval of parameter fluctuation period, the feature description of parameter change trend, and the interval of parameter correlation.

[0033] For continuous operation mode identifiers, the bound parameter fluctuation period range is the period range of continuous operation statistically analyzed from historical data. The parameter change trend is characterized by linear change, and the parameter correlation range is a high correlation range. For example, for the continuous operation mode identifier of an air conditioning system, its parameter fluctuation period range may be a fixed duration range, the change trend is linear, and the correlation range is a high value range.

[0034] The parameter fluctuation period range bound to the intermittent operation mode identifier is the period range during intermittent operation, with a step-like trend and a low parameter correlation range. For example, the intermittent operation mode identifier for a conference room lighting system has a period range that is not fixed and has an intermittent duration, with a step-like trend and a low correlation range.

[0035] The parameter fluctuation cycle range bound to the alternating operation mode identifier is a range that alternates between continuous and intermittent mode ranges, with a periodic switching trend. The parameter correlation range also alternates between high and low. For example, the alternating operation mode identifier of an elevator system has a cycle range that covers both continuous and intermittent cycles, with a periodic switching trend and a correlation range that also alternates between high and low.

[0036] Step S113: Extract resource usage-related parameters from the running status parameter stream and calculate resource usage characteristics. The resource usage characteristics include resource allocation ratio, resource switching frequency, and resource idle time. The resource allocation ratio is determined based on the ratio of resource consumption parameters to total resource parameters in the parameter stream.

[0037] For air conditioning systems, resource utilization parameters include compressor power consumption and fan power consumption. Resource allocation ratio refers to the ratio of the air conditioning system's power consumption at a given moment to the total resource parameters allocated to the air conditioning system within the building's total power supply. For example, at a given moment, the air conditioning system's power consumption is a fixed value, and the total resource parameter is the maximum power supply available to the system; the ratio of these two values ​​is the resource allocation ratio at that time.

[0038] Resource switching frequency refers to the ratio of the number of times parameters such as compressor operating frequency and fan speed are switched during the operation of an air conditioning system to the time. For example, if the compressor operating frequency is switched a certain number of times in one hour, the resource switching frequency is obtained by dividing the number of switches by one hour.

[0039] Resource idle time refers to the time during which certain components in an air conditioning system are idle. For example, at night, the air conditioning in some areas may stop operating, and the idle time is the resource idle time.

[0040] For lighting systems, resource usage parameters are the power consumption of each luminaire. Resource allocation ratio refers to the ratio of the total power consumption of the lighting system to the total power supplied to the lighting system. Resource switching frequency refers to the ratio of the number of times the luminaire's on / off state and brightness value are switched to the time. Resource idle time refers to the time the luminaire is in the off state.

[0041] Elevator system resource utilization parameters include the power consumption of elevator operation. Resource allocation ratio refers to the ratio of the elevator system's power consumption to the total power supplied to the elevator system. Resource switching frequency refers to the ratio of the number of times parameters such as elevator direction and speed are switched to the time interval. Resource idle time refers to the time the elevator is in standby mode.

[0042] Resource usage parameters for security systems include the power consumption of surveillance cameras, access control systems, etc. Resource allocation ratio refers to the ratio of the total power consumption of the security system to the total power supplied to the security system. Resource switching frequency refers to the ratio of the number of times the operating mode of surveillance cameras, the on / off status of access control systems, etc., is switched to the time interval. Resource idle time refers to the time that some security devices are in a dormant state.

[0043] Step S1114: Capture the response parameters of each subsystem during the parameter interaction process and generate interaction response features. The interaction response features include response delay features, response stability features and response correlation features. The response delay features are determined based on the interval parameter between parameter sending and receiving.

[0044] Within the office building's subsystems, there are instances of parameter interaction. For example, if a security system detects someone entering a certain area, it can send this parameter to the lighting and air conditioning systems. The response parameters during these interactions are captured to generate interaction response characteristics.

[0045] Response delay characteristics refer to the time interval between a subsystem sending a parameter and another subsystem receiving that parameter and responding. For example, the time interval between a security system sending a signal that someone has entered and a lighting system turning on the lights in that area is the response delay characteristic of the lighting system relative to the security system. By measuring these time intervals multiple times, multi-dimensional response delay characteristic data is generated.

[0046] Response stability characteristics refer to the degree of stability of a subsystem's response after receiving interactive parameters. For example, when the same person entry signal is received multiple times, whether the lighting system's turn-on time remains stable within a certain range; if the time fluctuation is small, the response stability characteristics are good.

[0047] Response correlation characteristics refer to the degree of correlation between different subsystems during the response process. For example, when a security system sends a signal that someone has entered, are there any correlations between the responses of the lighting system and the air conditioning system? After the lighting system is turned on, does the air conditioning system adjust the temperature accordingly? The tightness of these correlations constitutes the response correlation characteristics.

[0048] Step S115: The operating mode features, resource usage features, and interactive response features are classified and integrated according to subsystems to form a target state feature set. The features of each subsystem correspond to independent feature items in the target state feature set.

[0049] The operational mode characteristics, resource consumption characteristics, and interaction response characteristics of each subsystem extracted earlier are classified into air conditioning system, lighting system, elevator system, and security system. The three types of characteristics of each subsystem together constitute the independent feature terms of that subsystem in the target state feature set.

[0050] For example, the characteristic items of an air conditioning system include its continuous operation mode identifier and corresponding characteristic range, resource allocation ratio, resource switching frequency, resource idle time, response delay characteristics, response stability characteristics, and response correlation characteristics; the characteristic items of a lighting system include its intermittent operation mode identifier and corresponding characteristic range, etc. The characteristic items of all subsystems are integrated together to form a complete set of target state characteristics.

[0051] Step S120: Match the target state feature set with the preset collaborative control constraints to generate a collaborative correlation coefficient set between subsystems. The collaborative correlation coefficient set is used to characterize the degree of mutual influence between different subsystems during operation.

[0052] The preset collaborative control constraints are pre-defined based on the office building's operational management needs and the characteristics of each subsystem, including the rules and restrictions that each subsystem must follow when operating collaboratively. For example, under the principle of energy conservation first, the operation of the air conditioning system and the lighting system needs to cooperate with each other to avoid unnecessary energy waste; the operation of the elevator system needs to be linked with the security system to ensure the safety of personnel transportation.

[0053] Matching the target state feature set with these constraints involves checking whether the current state features of each subsystem meet the requirements of the constraints, and, if so, how much mutual influence exists between the subsystems. Through this analysis, synergistic correlation coefficients characterizing the degree of mutual influence are calculated, thus forming a set of synergistic correlation coefficients.

[0054] Step S121: Analyze the preset collaborative control constraints, extract the associated benchmark features and influence threshold ranges from the constraints, and the associated benchmark features include subsystem type matching features, operating parameter compatibility features and resource sharing features.

[0055] The pre-defined collaborative control constraints are analyzed in detail. For example, the constraints may stipulate that "the air conditioning system and the lighting system should be coordinated and adjusted according to the distribution of personnel during operation." This condition is analyzed to extract the associated baseline features and the range of influence thresholds.

[0056] Subsystem type matching characteristics refer to the characteristics of whether different types of subsystems are suitable for collaborative operation. For example, air conditioning systems and lighting systems both belong to the category of environmental regulation subsystems in buildings, and their type matching degree is relatively high. Although elevator systems and security systems have different functions, they are related in terms of personnel management and also have certain type matching characteristics.

[0057] Operating parameter compatibility refers to whether the operating parameters of each subsystem are compatible when operating together. For example, the temperature adjustment range of the air conditioning system and the brightness adjustment range of the lighting system will not conflict when operating together.

[0058] Resource sharing characteristics refer to whether there is a sharing and allocation relationship in the use of resources among various subsystems. For example, air conditioning systems and lighting systems both consume electricity, and when power supply resources are limited, there is a problem of resource sharing and allocation.

[0059] The influence threshold range refers to the range of values ​​for the synergistic correlation coefficient. When the correlation coefficient is within a certain range, the synergistic operation between subsystems is in a reasonable state.

[0060] Step S1211: The preset collaborative control constraints are decomposed into condition items, feature items, and threshold items. The condition items include the preconditions for collaborative operation, the feature items include the associated benchmark features, and the threshold items include the range of influencing thresholds.

[0061] For example, the constraint "When the personnel density in the office building exceeds a certain value, the elevator system should increase its operating frequency, and the security system should strengthen its monitoring" can be structurally broken down. The condition is "the personnel density in the office building exceeds a certain value"; the feature items include the type matching characteristics of the elevator system and the security system (both are related to personnel management), the compatibility characteristics of operating parameters (the compatibility between the elevator operating frequency and the security monitoring frequency), and the resource sharing characteristics (the sharing of power resources between the two); the threshold items include the adjustment threshold of the elevator operating frequency, the threshold of the security monitoring intensity, and the threshold range of the synergistic correlation coefficient between the two.

[0062] Step S1212: Extract subsystem type matching features from the feature items. The subsystem type matching features include type compatibility identifiers, functional complementarity identifiers, and hierarchical correspondence identifiers. The type compatibility identifier is used to indicate whether different types of subsystems can operate collaboratively.

[0063] From the above-described feature items, subsystem type matching features are extracted. The type compatibility identifier is used to indicate whether different subsystems can operate together. For example, air conditioning systems and ventilation systems are type compatible and can operate together, so the type compatibility identifier is "compatible"; while elevator systems and fire protection systems may have operational conflicts in some cases, so the type compatibility identifier may be "limited compatibility".

[0064] The complementary function indicator refers to whether the functions of different subsystems can complement each other. For example, the personnel detection function of a security system and the area lighting function of a lighting system are complementary, and the complementary function indicator is "complementary". On the other hand, two independent lighting systems with similar functions are identified as "weakly complementary".

[0065] Hierarchical correspondence refers to the correspondence between different subsystems in the building management hierarchy. For example, there is a hierarchical correspondence between the central control system and each subsystem, and the hierarchical correspondence is indicated by "superior-subordinate correspondence"; while the subsystems belong to the same level, and the hierarchical correspondence is indicated by "same level correspondence".

[0066] Step S1213: Extract the compatibility features of the operating parameters. The compatibility features of the operating parameters include the overlap of parameter ranges, the consistency of parameter change trends, and the adaptability of parameter units. The overlap of parameter ranges is determined based on the overlapping part of the value range of the operating parameters of each subsystem.

[0067] Taking air conditioning and ventilation systems as examples, we extract their operational parameter compatibility characteristics. Parameter range overlap refers to the overlapping portion of the temperature regulation range (e.g., 18-26 degrees Celsius) of the air conditioning system and the airflow regulation range (corresponding to different temperature regulation effects) of the ventilation system. The parameter range overlap is obtained by calculating the ratio of the length of the overlapping interval to the total interval length.

[0068] Consistency of parameter change trends refers to whether the temperature change trend of the air conditioning system and the air volume change trend of the ventilation system are consistent. For example, when the indoor temperature rises, the air volume of the ventilation system also increases. If the two change trends are consistent, then the consistency of parameter change trends is high.

[0069] Parameter unit compatibility refers to whether the units of the operating parameters of two subsystems can be compatible with each other to achieve effective data interaction and collaborative control. For example, the temperature unit of an air conditioning system is degrees Celsius, and the air volume unit of a ventilation system is cubic meters per hour. Although the units are different, in the collaborative control logic, the parameter changes of the two can be correlated through preset conversion rules, so the parameter unit compatibility is high.

[0070] For elevator and security systems, the extraction of operational parameter compatibility features follows the same logic. Parameter range overlap is reflected in the overlap between the elevator's load capacity and the security system's allowed personnel capacity; this feature value is determined by calculating the percentage of the overlapping area. Parameter change trend consistency is demonstrated by whether the elevator's operating frequency increases when the security system detects an increase in personnel flow; if both show a synchronous trend, the feature value is higher. Regarding parameter unit adaptability, the elevator's operating unit is a floor, while the security system's personnel counting unit is a person. By establishing a correspondence between floors and personnel distribution, unit adaptability is achieved, ensuring that parameters can be correctly interpreted and used during collaborative control.

[0071] Step S1214: Extract resource sharing features, which include resource type matching degree, resource allocation conflict rate and resource reuse efficiency. The resource type matching degree is determined based on the number of overlapping categories of resources required by the subsystem.

[0072] Taking air conditioning and lighting systems as examples, both rely on electricity as their operating resource, and their resource types overlap. By statistically analyzing the categories of resources required by both systems, such as electricity and control system interfaces, and calculating the ratio of the number of overlapping categories to the total number of resource categories, we can obtain the resource type matching degree. A higher resource type matching degree is indicated by a greater overlap in the types of resources required by both systems.

[0073] Resource allocation conflict rate refers to the probability that a conflict will occur between two subsystems during resource allocation when resources are limited. For example, during peak electricity consumption periods, both air conditioning and lighting systems require a significant amount of electricity. If the total power supply cannot meet the maximum demand of both, a resource allocation conflict may occur. The resource allocation conflict rate is determined by statistically analyzing the ratio of the number of conflicts that occurred during historical operation to the total operating time.

[0074] Resource reuse efficiency refers to the degree to which resources used in one subsystem can be reused in another subsystem. For example, the air ducts of an air conditioning system can be reused by a ventilation system, reducing redundant construction and waste of resources. This characteristic value is determined by calculating the utilization rate of reusable resources.

[0075] For elevator and fire protection systems, the extraction of resource sharing characteristics needs to consider resource coordination in emergency situations. Resource type matching is reflected in the overlap of resource categories such as emergency power supplies and communication lines; the greater the overlap, the higher the characteristic value. Resource allocation conflict rate mainly occurs in emergency situations, specifically whether there is resource competition between the emergency operation of the elevator and the power supply to the fire extinguishing equipment of the fire protection system. This characteristic value is calculated by analyzing conflicts in historical emergency events. Resource reuse efficiency is reflected in whether the elevator car can be reused as a fire evacuation tool in emergency situations; this characteristic value is determined based on the frequency and effectiveness of reuse.

[0076] Step S1215: Extract the influence threshold range from the threshold items. The influence threshold range includes the lower limit of the correlation coefficient, the upper limit of the correlation coefficient, and the critical transformation value. Each correlation benchmark feature corresponds to a set of influence threshold ranges.

[0077] The threshold range for subsystem type matching characteristics is set based on historical collaborative operation data. The lower limit of the correlation coefficient refers to the value at which the subsystem type matching degree reaches, indicating that the two systems have the basis for collaborative operation. The upper limit of the correlation coefficient refers to the highest value of the type matching degree, at which point the potential for collaborative operation is greatest. The critical transition value refers to the value at which the collaborative operation mode will change, for example, from basic collaboration to deep collaboration.

[0078] For the compatibility characteristics of operating parameters, the range of influence thresholds is determined based on the effectiveness of parameter coordination. The lower limit of the correlation coefficient ensures that parameters can at least perform basic interactions; the upper limit of the correlation coefficient corresponds to the ideal state of complete parameter compatibility; the critical transition value is the critical point at which the parameter compatibility state undergoes a qualitative change. Beyond this value, parameter coordination can achieve a more efficient control effect.

[0079] The impact threshold range of resource sharing characteristics is related to resource utilization efficiency. The lower limit of the correlation coefficient ensures that resource sharing will not have a negative impact; the upper limit of the correlation coefficient represents the optimal state of resource sharing; the critical transition value is the turning point from effective to inefficient resource sharing. When the correlation coefficient falls below this value, the resource sharing strategy needs to be adjusted.

[0080] Step S122: Compare each subsystem feature in the target state feature set with the associated benchmark feature item by item, and calculate the feature matching degree. The feature matching degree is determined based on the overlap and similarity of feature items.

[0081] Taking the comparison between the air conditioning system features in the target state feature set and the subsystem type matching features in the associated baseline features as an example, the process first checks whether the type identifier of the air conditioning system overlaps with the type specified in the associated baseline features. If both belong to the environmental regulation subsystem category, the overlap is counted as a certain value. Next, the similarity between the two in terms of functional description, hierarchical relationship, etc., is analyzed. If the similarity is high, the similarity score is high. The overlap and similarity scores are combined and calculated according to preset weights to obtain the matching degree of this feature item.

[0082] For the comparison of compatibility features of operating parameters, the temperature adjustment range and trend of the air conditioning system are compared item by item with the corresponding parameter range and trend requirements in the associated benchmark features. The percentage of overlap between the actual parameters and the benchmark parameters is calculated as the degree of overlap, and the similarity is determined by the degree of match between the parameter change curves. The combination of the two results in the degree of matching of the compatibility features of operating parameters.

[0083] The comparison of resource sharing features also follows the above process. The resource requirements and allocation of the air conditioning system are compared with the resource sharing standards in the associated benchmark features. The overlap of resource categories and the similarity of resource utilization patterns are calculated to obtain the matching degree of resource sharing features.

[0084] Other subsystems, such as the lighting system, are compared with the associated baseline features using the same logic to ensure that each feature item receives a corresponding feature matching degree.

[0085] Step S123: Calculate the initial correlation coefficient based on the feature matching degree and the influence threshold range. When the feature matching degree is in the upper limit range of the influence threshold range, the initial correlation coefficient takes the corresponding high value. When the feature matching degree is in the lower limit range of the influence threshold range, the initial correlation coefficient takes the corresponding low value.

[0086] For example, if the feature matching degree between the air conditioning system and the lighting system is in the upper limit of the influence threshold range, it indicates that the two perform well in terms of type matching, parameter compatibility and resource sharing. The initial correlation coefficient is taken as the high value in this range to reflect the strong mutual influence between the two.

[0087] When the feature matching degree is in the middle range of the influence threshold, the initial correlation coefficient is determined linearly based on the specific position of the matching degree within the range. For example, if the influence threshold range is 0.3-0.7 and the feature matching degree is 0.5, then the initial correlation coefficient is taken as the middle value corresponding to 0.5.

[0088] If the feature matching degree is lower than the lower limit of the influence threshold range, it indicates that the foundation for the two to work together is poor. The initial correlation coefficient should be set to a corresponding low value, indicating that the degree of mutual influence is weak.

[0089] Step S124: Perform cross-validation on the initial correlation coefficient. The validation process is based on the actual impact data of different subsystems in historical collaborative operation, and adjust the initial correlation coefficient that deviates from the actual impact.

[0090] Collect historical collaborative operation data of the air conditioning and lighting systems over the past year, including their operating status parameters and collaborative control effects. Analyze the response and impact of the lighting system when the air conditioning system is adjusted, and calculate the actual mutual influence coefficient. Compare the initial correlation coefficient with the actual influence coefficient; if the deviation is significant, adjust the initial correlation coefficient based on the actual influence coefficient.

[0091] For example, if the initial correlation coefficient is calculated to be 0.6, but historical data shows an actual impact coefficient of 0.4, this indicates that the initial correlation coefficient is too high and needs to be adjusted downwards to 0.45, which is closer to the actual value. During the adjustment process, the reliability and timeliness of historical data must be considered, and recent data without abnormal interference should be prioritized as the basis for verification.

[0092] The initial correlation coefficient between the elevator system and the security system was also cross-validated by analyzing historical cases of their coordinated operation, such as the actual impact of changes in elevator operating frequency on security monitoring pressure. The initial correlation coefficient was adjusted based on these actual data to ensure that it could truly reflect the degree of mutual influence between the subsystems.

[0093] Step S125: The verified and adjusted correlation coefficients are sorted and organized by subsystem pair to form a set of collaborative correlation coefficients. Each subsystem pair corresponds to a correlation value and a description of the correlation characteristics.

[0094] The verified and adjusted correlation coefficients between the air conditioning system and the lighting system, and between the air conditioning system and the elevator system, are categorized according to subsystem pairs. For example, the subsystem pair of "air conditioning system-lighting system" corresponds to a correlation value, along with a description of the correlation characteristics, such as "the two are closely related in terms of resource sharing, and moderately related in terms of parameter compatibility."

[0095] The correlation coefficients of the "elevator system - security system" subsystem pairs are also organized in the same way. The correlation characteristics description may include "strong correlation in terms of personnel flow regulation, and weak correlation in terms of resource type matching," etc. The correlation coefficients and correlation characteristics descriptions of all subsystem pairs are integrated together to form a complete set of collaborative correlation coefficients.

[0096] Step S130: Construct a collaborative adjustment target model based on the collaborative correlation coefficient set and the target state feature set. The variables of the collaborative adjustment target model include the adjustment amount of the operating parameters and the adjustment time sequence parameters of each subsystem.

[0097] The construction of the collaborative regulation target model aims to achieve the coordinated and optimized operation of each subsystem. The set of collaborative correlation coefficients serves as the weighting basis for the mutual influence between subsystems. Combined with the current state of each subsystem in the target state feature set, the model's variables and objective function are determined.

[0098] Operating parameter adjustment amounts refer to the specific numerical changes in operating parameters that each subsystem needs to adjust in order to achieve collaborative optimization goals, such as temperature adjustment for an air conditioning system or brightness adjustment for a lighting system. Adjustment timing parameters refer to the order and time intervals in which each subsystem performs adjustment operations, such as adjusting the operating frequency of an elevator system first, and then adjusting the monitoring range of a security system.

[0099] When building a model, it is necessary to clarify the relationships between these variables and how they affect the goals of coordinated regulation, such as energy efficiency and operational stability.

[0100] Step S130 specifically includes: Step S131: Take each correlation coefficient in the collaborative correlation coefficient set as a weighting factor and multiply it with the interaction response feature in the corresponding target state feature set to generate the correlation influence value between subsystems.

[0101] Taking the "air conditioning system - lighting system" subsystem pair as an example, the correlation coefficient of this subsystem pair in the collaborative correlation coefficient set is 0.5. The interaction response features between the air conditioning system and the lighting system in the target state feature set include response delay features, response stability features, and response correlation features, each of which is a multi-dimensional set of values. Multiplying the correlation coefficient 0.5 by the value of each dimension of these interaction response features yields a new set of multi-dimensional values, which is the correlation influence value of this subsystem pair.

[0102] For the "elevator system - security system" subsystem pair, its correlation coefficient is also multiplied by the corresponding interaction response feature. For example, if the correlation coefficient is 0.6, and the response correlation feature in the interaction response feature is a set of values ​​representing the degree of correlation between the two responses, the resulting correlation impact value can reflect the degree of mutual influence after considering the weights.

[0103] The combined influence values ​​of all subsystem pairs constitute the part of the model that characterizes the interactions between subsystems.

[0104] Step S132: Using the adjustment amount of the operating parameters of each subsystem as variables, and combining the associated influence value, construct the first part of the collaborative adjustment target model, which is used to characterize the sum of the associated influence after the subsystem is adjusted.

[0105] The operating parameters of each subsystem are set as variables. For example, the temperature adjustment of the air conditioning system is set as variable A, the brightness adjustment of the lighting system is set as variable B, the operating frequency adjustment of the elevator system is set as variable C, and the monitoring range adjustment of the security system is set as variable D.

[0106] These variables are combined with their corresponding associated impact values ​​to construct the first part of the model. For example, variable A is associated with the associated impact values ​​of subsystem pairs such as "air conditioning system-lighting system" and "air conditioning system-elevator system". By integrating them through a pre-defined functional relationship, a comprehensive expression is obtained. The result of this expression is the sum of the associated impacts after the subsystems have been adjusted.

[0107] The construction of this part of the model needs to ensure that the relationship between variables and associated impact values ​​can accurately reflect the effect of changes in the moderating amount on the mutual influence between subsystems. For example, how does the sum of its associated impact with other subsystems change when variable A increases?

[0108] Step S133: Extract resource occupancy features from the target state feature set, multiply the resource occupancy features by the adjustment time series parameters, and generate a feature quantity of resource consumption changing over time, which serves as the second part of the collaborative adjustment target model.

[0109] Resource occupancy characteristics of each subsystem are extracted from the target state feature set, such as the resource allocation ratio and resource switching frequency of the air conditioning system, and the idle time of the lighting system. Adjustment timing parameters include the start time and duration of adjustment operations for each subsystem.

[0110] Multiplying the resource occupancy characteristics by the adjustment time series parameters—for example, multiplying the resource allocation ratio of the air conditioning system by the duration in the adjustment time series parameters—gives the resource consumption characteristics of the air conditioning system during the adjustment period; the idle time of the lighting system is correlated with the start time in the adjustment time series parameters to calculate the changes in resource consumption in different time periods.

[0111] These calculated characteristics of resource consumption over time are integrated to form the second part of the coordinated adjustment target model, which is used to characterize the dynamic changes in resource consumption under different adjustment time series.

[0112] Step S134: Set the constraints of the coordinated adjustment target model. The constraints include the range of values ​​of the operation parameter adjustment amount, the constraints of the order of adjustment time parameters, and the upper limit constraints of the associated influence values. Each constraint corresponds to a set of boundary conditions determined based on historical data.

[0113] The range of values ​​for operating parameter adjustments is determined based on the safety operation requirements and performance limitations of each subsystem. For example, the temperature adjustment of an air conditioning system cannot exceed its maximum adjustment range; otherwise, it may lead to equipment damage or malfunction. By analyzing the parameter adjustment range during normal operation of the equipment in historical data, upper and lower limits for operating parameter adjustments are set as constraint boundaries.

[0114] The order of adjustment timing parameters is determined based on the logical dependencies between subsystems. For example, after a security system detects personnel entering an area, it should first turn on the lighting system in that area, and then adjust the temperature of the air conditioning system. Therefore, the adjustment timing parameters of the lighting system should be earlier than those of the air conditioning system. Based on successful historical cases of coordinated control, the order of adjustment timing for each subsystem is determined.

[0115] The upper limit constraint on the correlation impact value is to prevent excessive mutual influence between subsystems from causing system instability. For example, the correlation impact value between the air conditioning system and the ventilation system cannot exceed a certain upper limit; otherwise, problems such as excessive airflow or sudden temperature changes may occur. This upper limit constraint is set based on the range of correlation impact values ​​during the system's stable operation in history.

[0116] Step S1341: Extract the parameter fluctuation range of each subsystem under safe operating conditions from historical operating data, and use the extreme value of the parameter fluctuation range as the boundary of the value range of the operating parameter adjustment amount to form the first constraint sub-item.

[0117] For air conditioning systems, analyze the fluctuation data of parameters such as return air temperature and supply air volume under safe operating conditions over the past year to determine the maximum and minimum values ​​of these fluctuations. Use these extreme values ​​as the upper and lower boundaries of the air conditioning system's operating parameter adjustment quantities. For example, the upper limit of the return air temperature adjustment quantity is one value, and the lower limit is another value, forming part of the first constraint sub-item.

[0118] Similarly, for lighting systems, the extreme fluctuations of parameters such as brightness and power consumption during safe operation are extracted as the boundaries of the adjustment range of the lighting system's operating parameters; for elevator systems, the extreme fluctuations of parameters such as operating speed and load capacity are used as the boundaries of their operating parameter adjustment ranges; and for security systems, the extreme fluctuations of parameters such as monitoring range and alarm sensitivity are used as their corresponding boundaries. These boundaries of all subsystems together constitute the first constraint item.

[0119] Step S1342: Analyze the correlation data between the adjustment execution order and adjustment effect of each subsystem in the historical coordinated adjustment process, determine the rules of the order relationship of adjustment timing parameters, and when the adjustment result of one subsystem affects the adjustment basis of another subsystem, the adjustment timing parameters of the first subsystem must be earlier than the adjustment timing parameters of the second subsystem, forming the second constraint item.

[0120] Analysis of historical data on coordinated adjustments between elevator and security systems reveals that the best adjustment results are achieved when the security system clears personnel from a specific area before the elevator system shuts down that area. Reversing this sequence can lead to problems such as people becoming trapped. Therefore, it is determined that when the security system's adjustment result (personnel clearance) affects the elevator system's adjustment basis (area status), the security system's adjustment timing parameters must precede those of the elevator system.

[0121] Historical data shows that adjusting the airflow of the ventilation system before adjusting the temperature of the air conditioning system can more efficiently regulate the indoor environment. Therefore, the timing parameters for adjusting the ventilation system should precede those for adjusting the air conditioning system. These rules of sequence constitute the second constraint sub-item.

[0122] Step S1343: Calculate the maximum safe threshold of the associated impact value of each subsystem in the historical operation. When the associated impact value exceeds the maximum safe threshold, it will cause an operational abnormality. Use the maximum safe threshold as the upper limit constraint of the associated impact value to form the third constraint sub-item.

[0123] The correlation between the air conditioning system and the lighting system was statistically analyzed during historical operation. When this value exceeded a certain threshold, abnormal operational situations such as excessive air conditioning load leading to circuit breaker tripping and unstable lighting system voltage occurred. This threshold was determined as the maximum safe threshold for the correlation between the two systems, serving as an upper limit constraint.

[0124] Similarly, for elevator and fire protection systems, operational anomalies that occurred when the historical correlation impact value exceeded a certain threshold are statistically analyzed, such as elevator stoppages caused by conflicts between elevator emergency operation and fire alarm signals. This threshold is used as the upper limit constraint on the correlation impact value between the two systems. The upper limit constraint on the correlation impact value of all subsystem pairs constitutes the third constraint item.

[0125] Step S1344: Set the adjustment coefficient of the boundary condition for each constraint sub-item. The adjustment coefficient is determined based on the operating mode features in the current target state feature set. The adjustment coefficient corresponding to the continuous operating mode is less than the adjustment coefficient corresponding to the intermittent operating mode.

[0126] For the first constraint, when the air conditioning system is in continuous operation mode, the range of its operating parameter adjustment is relatively fixed, and the adjustment coefficient is set to a small value; when it is in intermittent operation mode, due to the large changes in the operating state, greater adjustment flexibility is required, and the adjustment coefficient is set to a larger value so that the range of values ​​can be appropriately widened.

[0127] The adjustment coefficient for the second constraint item is also determined based on the characteristics of the operating mode. When the elevator system is in continuous operation mode, the order of adjustment timing is more strictly controlled, and the adjustment coefficient is smaller to ensure the rigid execution of the timing rules; when in intermittent operation mode, the adjustment coefficient is larger, allowing for flexible adjustment of the timing sequence within a certain range.

[0128] The adjustment coefficient of the third constraint item also follows this logic. In continuous operation mode, the subsystem operates more stably, and the upper limit constraint adjustment coefficient of the associated influence value is smaller and the restriction is stricter. In intermittent operation mode, the adjustment coefficient is larger and the upper limit constraint can be appropriately relaxed.

[0129] Step S1345: Integrate the first constraint sub-item, the second constraint sub-item, the third constraint sub-item and the corresponding adjustment coefficient into the constraint terms of the collaborative adjustment target model. The constraint terms are incorporated into the collaborative adjustment target model in the form of inequalities.

[0130] The range of adjustment values ​​for each subsystem's operating parameters in the first constraint sub-item is expressed using inequalities. For example, the temperature adjustment for the air conditioning system is greater than or equal to the lower limit and less than or equal to the upper limit, multiplied by the corresponding adjustment coefficient. The timing relationship of adjustments in the second constraint sub-item is transformed into the magnitude relationship of timing parameters using inequalities. For example, the timing parameter for the security system is less than that for the elevator system. The upper limit constraint of the associated influence value in the third constraint sub-item is expressed as the associated influence value being less than or equal to the maximum safety threshold multiplied by the adjustment coefficient.

[0131] The aforementioned inequalities collectively constitute the set of constraints for the coordinated adjustment target model. Each inequality corresponds to a specific constraint condition, ensuring that the model does not exceed the boundaries of safe operation and coordinated cooperation of the subsystems during the solution process. For example, regarding the coordinated adjustment of the air conditioning system and the ventilation system, the first constraint inequality limits the fluctuation range of the air conditioning temperature adjustment and the ventilation air volume adjustment, preventing equipment damage or energy waste due to excessive adjustment amplitude; the second constraint inequality clarifies that when the security system detects a densely populated area, the ventilation system's adjustment sequence must take precedence over the air conditioning system to quickly achieve air circulation; and the third constraint inequality controls the correlation influence value between the air conditioning and ventilation systems, preventing their coordinated adjustment from causing disturbances to the indoor environment beyond the safe range.

[0132] These constraints are logically integrated into the collaborative adjustment target model, forming a complete constraint system. During model construction, the coefficients and boundary values ​​of each inequality can be verified to ensure they match the operating mode characteristics in the target state feature set. For example, when the air conditioning system is in continuous operation mode, its corresponding adjustment coefficient is smaller, and the boundaries of the constraints are relatively loose to meet the requirements of stable operation; while when the air conditioning system is in intermittent operation mode, the adjustment coefficient is larger, and the boundaries of the constraints are more stringent to avoid frequent start-stop operations causing damage to the equipment.

[0133] Step S135: Integrate the first part, the second part, and the constraint terms into a collaborative adjustment target model. The output value of the collaborative adjustment target model is the comprehensive adjustment benefit value, which corresponds to the sum of related impacts and resource consumption characteristics.

[0134] After constructing the first part, the second part, and the constraint terms, these three parts are integrated into a synergistic adjustment target model. The first part represents the sum of the interrelated effects of the subsystems after adjustment, and the second part represents the characteristic quantity of resource consumption changing over time. Together, they serve as the core inputs of the model and are transformed into a comprehensive adjustment benefit value through a defined mapping relationship. The comprehensive adjustment benefit value is a multi-dimensional feature vector, where each dimension corresponds to the synergistic benefit of different subsystem combinations, such as the energy-saving benefit dimension of air conditioning and lighting systems, and the operational efficiency dimension of elevators and security systems.

[0135] The comprehensive adjustment benefit value is positively correlated with the sum of related impacts; that is, the more reasonable and synergistic the related impacts among subsystems, the larger the sum of related impacts, and the higher the dimensional value corresponding to the comprehensive adjustment benefit value. Simultaneously, the comprehensive adjustment benefit value is negatively correlated with resource consumption characteristics; the smaller the resource consumption characteristics, the more efficient the resource utilization, and the higher the dimensional value corresponding to the comprehensive adjustment benefit value. Constraint terms limit the solution space of the model through inequalities, ensuring that the calculation of the comprehensive adjustment benefit value is always within a reasonable constraint range.

[0136] For example, in the coordinated adjustment during peak office hours, the first part of the model calculates the total correlation impact between air conditioning, elevators, and security systems, while the second part statistically analyzes the resource consumption characteristics of these systems. These two parts are then transformed into various dimensions of the overall adjustment benefit value through a pre-defined mapping rule. If the total correlation impact is high and the resource consumption characteristics are low, the overall adjustment benefit value will be high, indicating that the current adjustment scheme is relatively optimized.

[0137] Step S140: Solve the collaborative adjustment target model to obtain the optimal combination of adjustment parameters that satisfies the constraints. The optimal combination of adjustment parameters includes the specific adjustment value and execution order of each subsystem.

[0138] The process of solving the coordinated regulation target model involves finding the optimal combination of operational parameter adjustments and timing parameters within the solution space defined by constraints. This process requires iteratively approaching the optimal solution by combining the target state feature set and the coordinated correlation coefficient set.

[0139] Before starting the solution process, the initial search range can be determined based on the number of subsystems and the dimensions of parameters to ensure that the search covers all possible effective adjustment schemes. During the solution process, each adjustment of parameter combinations must satisfy all inequalities in the constraint terms to avoid invalid solutions or solutions that exceed the safe range. By continuously optimizing the parameter combinations, the overall adjustment benefit value is continuously improved until a stable optimal solution is found.

[0140] Step S141: The collaborative adjustment target model is solved using the feature space search method. The adjustment amount of the running parameters and the adjustment time sequence parameters are used as the dimensions of the search space, and the value range of each dimension corresponds to the boundary conditions in the constraint terms.

[0141] The feature space search method constructs a multi-dimensional search space using the operating parameter adjustment amount and the adjustment timing parameter as coordinate axes. The value range of each coordinate axis is determined by the boundary conditions in the constraint terms. For example, the temperature adjustment amount of the air conditioning system is one dimension, and its value range corresponds to the upper and lower limits of the temperature adjustment amount in the first constraint term; the adjustment timing parameter of the elevator system is another dimension, and its value range is limited by the timing sequence relationship in the second constraint term.

[0142] In the search space, each point represents a combination of operational parameter adjustments and timing parameters. By systematically searching the space and traversing possible parameter combinations, a foundation is laid for subsequent selection of the optimal solution. During the search, different search precisions can be set according to the importance of each dimension. For parameter dimensions that have a greater impact on the overall adjustment benefit value, a higher search precision is used to improve the accuracy of the solution.

[0143] Step S142: Generate initial parameter combination samples within the search space. The number of initial parameter combination samples is determined based on the number of subsystems and parameter dimensions. Each initial parameter combination sample contains a set of operating parameter adjustment amounts and adjustment timing parameters.

[0144] The initial parameter combination samples are generated using a uniform sampling method, selecting sample points evenly across all dimensions of the search space to ensure a uniform distribution of samples and avoid missing optimal solutions due to samples being concentrated in a certain area. Determining the sample size requires comprehensive consideration of the number of subsystems and the parameter dimensions; the more subsystems and the higher the parameter dimensions, the larger the initial sample size should be to ensure representativeness.

[0145] For example, when four subsystems are involved—air conditioning, lighting, elevators, and security—each containing two operational parameter adjustments and one timing parameter, the parameter dimension is 12-dimensional. The initial number of parameter combination samples is determined based on the complexity of this 12-dimensional space, typically selecting a sample size sufficient to cover the key value ranges of each dimension. Within each sample, the temperature adjustment for the air conditioning system, the brightness adjustment for the lighting system, the operating speed adjustment for the elevator system, the monitoring frequency adjustment for the security system, and their respective timing parameters all have specific values.

[0146] Step S143: Calculate the output value of the collaborative adjustment target model corresponding to each initial parameter combination sample, and select the samples whose output values ​​meet the preset optimization target as candidate samples.

[0147] For each initial parameter combination sample, it is substituted into the collaborative adjustment target model to calculate the corresponding comprehensive adjustment benefit value. The preset optimization target is set according to the needs of building intelligence, such as all dimensions of the comprehensive adjustment benefit value reaching or exceeding the preset threshold, or the comprehensive adjustment benefit value being within a certain proportion range among all samples.

[0148] By comparing the comprehensive adjustment benefit value of each sample with the preset optimization objective, samples that meet the requirements are selected as candidate samples. This step can reduce the computational load of subsequent iterative optimizations and improve solution efficiency. For example, among the 1000 initial samples generated, the top 200 samples with the highest comprehensive adjustment benefit values ​​are selected as candidate samples.

[0149] Step S144: Iteratively optimize the candidate samples by adjusting the values ​​of the running parameters and the order of the time-series parameters to make the output value of the collaborative adjustment target model change towards a better direction. All conditions of the constraint terms must be met during the iteration process.

[0150] The iterative optimization process employs a stepwise approximation approach, fine-tuning the adjustment amounts and timing parameters of the candidate samples. After each adjustment, the overall adjustment benefit value is recalculated. If the benefit value improves and all constraints are met, the adjustment is retained; if the benefit value decreases or the constraints are violated, the adjustment is abandoned, and other adjustment directions are explored.

[0151] Iterative optimization is conducted in stages. First, the adjustment amount of the operating parameters is optimized, then the adjustment timing parameters are optimized, and this process is repeated until the overall adjustment benefit value no longer shows a significant improvement. During the optimization process, it is necessary to verify the satisfaction of the constraints in real time to ensure that each iteration step is carried out within a reasonable range.

[0152] For example, step S1441: Select the sample with the optimal output value of the collaborative adjustment target model from the candidate samples as the initial iteration point, and record the adjustment amount of the running parameters and the adjustment timing parameters of the sample.

[0153] Among the candidate samples, the sample with the highest comprehensive adjustment benefit value is selected as the initial iteration point. The adjustment amount of the operating parameters of each subsystem in this sample is recorded, such as the temperature adjustment amount of the air conditioning system is a certain value, the brightness adjustment amount of the lighting system is a certain value, etc., as well as the adjustment timing parameters of each subsystem, such as the adjustment timing parameter of the security system is t1, the adjustment timing parameter of the elevator system is t2, etc.

[0154] The selection of the initial iteration point provides a starting point for the subsequent optimization process, and its comprehensive adjustment benefit value serves as the benchmark for subsequent optimization. All subsequent adjustments aim to exceed this benchmark.

[0155] Step S1442: Based on the initial iteration point, adjust the values ​​of the operating parameters according to the preset step size, adjusting the adjustment amount of one subsystem each time, while keeping other parameters unchanged, and calculate the output value of the adjusted collaborative adjustment target model.

[0156] The preset step size is determined based on the parameter characteristics of the subsystem. For example, the preset step size for the temperature regulation of the air conditioning system is a certain value, and the preset step size for the brightness regulation of the lighting system is a certain value. Based on the initial iteration point, the temperature regulation of the air conditioning system is first adjusted by increasing the preset step size, while keeping the parameters of other subsystems unchanged, and the overall regulation benefit value after adjustment is calculated. Then, the temperature regulation of the air conditioning system is restored to its initial value, the brightness regulation of the lighting system is adjusted, and the overall regulation benefit value is calculated again.

[0157] By adjusting each subsystem individually as described above, the impact of the adjustment amount of each subsystem's operating parameter on the overall adjustment benefit value is determined, providing a basis for subsequent optimization directions.

[0158] Step S1443: If the adjusted output value is better than the original output value and meets the constraint conditions, then retain the adjustment and use the new parameter as the current iteration point; if not, revert to the original parameter and try to adjust the adjustment amount of other subsystems.

[0159] If, after adjusting the operating parameters of a subsystem, the new comprehensive adjustment benefit is higher than before the adjustment, and the new parameter satisfies all the inequalities in the constraint terms (e.g., the temperature adjustment is still within a safe range, and the timing relationship with other subsystems is not violated), then the adjustment is accepted, and the new parameter combination is used as the current iteration point.

[0160] If the adjusted output value is lower than before the adjustment, or if the constraint conditions are violated, then the adjustment is abandoned, the parameters are restored to their pre-adjustment state, and the adjustment of the operating parameters of the next subsystem continues.

[0161] Step S1444: After optimizing the running parameter adjustment amount to a local optimum, adjust the order of the adjustment timing parameters, swap the execution order of the two subsystems, calculate the output value of the adjusted collaborative adjustment target model, and determine whether the optimization direction and constraints are met.

[0162] When the overall adjustment benefit value no longer increases after multiple adjustments to the operating parameters, indicating a local optimum, the order of the adjustment timing parameters is then adjusted. For example, if the original adjustment timing parameters of the security system were less than those of the elevator system, their order can be swapped so that the adjustment timing parameters of the elevator system are less than those of the security system. The adjusted overall adjustment benefit value is then calculated.

[0163] Determine whether the new comprehensive adjustment benefit value is higher than that before the adjustment, and whether the adjusted time sequence relationship satisfies the inequality conditions in the constraint terms, such as whether it conforms to the sequential logic of the coordinated operation of subsystems.

[0164] Step S1445: Repeat the steps of adjusting the parameter adjustment amount and adjusting the adjustment timing parameters until the change in the output value of the collaborative adjustment target model is less than the preset threshold after K consecutive iterations. At this time, the determined parameter combination is the candidate sample after iterative optimization.

[0165] K is set to a certain integer, and a small preset threshold is used to determine whether the comprehensive adjustment benefit value tends to stabilize. In the process of K consecutive iterations, the adjustment amount of the operating parameters is adjusted first, and then the adjustment timing parameters are adjusted. If the change in the comprehensive adjustment benefit value after each iteration is less than the preset threshold, it indicates that the current parameter combination has reached a better state, the iteration is stopped, and the parameter combination is used as a candidate sample after iterative optimization.

[0166] For example, if the change in the comprehensive adjustment benefit value is less than the preset threshold after five consecutive iterations, the parameter combination at this time is the optimized candidate sample.

[0167] Step S145: When the output value of the coordinated adjustment target model after iterative optimization is stable within the preset range and no longer changes, stop the iteration, determine the corresponding parameter combination as the optimal adjustment parameter combination, and retain the parameter dimension corresponding to the target state feature set in the specific adjustment parameter combination. The execution order is determined based on the numerical value of the adjustment time sequence parameter.

[0168] During the iterative optimization process, if the overall adjustment benefit value remains stable within the preset optimal range and shows no significant change in multiple consecutive iterations, the optimal solution is considered to have been found, and the iteration stops. The parameter combination at this point is determined as the optimal adjustment parameter combination.

[0169] The specific adjustment values ​​in the optimal combination of adjustment parameters correspond one-to-one with the parameter dimensions of the target state feature set, ensuring the targeting and effectiveness of the adjustment. The execution order is determined by the magnitude of the adjustment timing parameters; the smaller the timing parameter value, the earlier it is executed. For example, if the adjustment timing parameter for a security system is t3 and the adjustment timing parameter for a lighting system is t4, and t3 is less than t4, the adjustment command for the security system will be executed first.

[0170] Step S150: The optimal combination of adjustment parameters is converted into a sequence of control instructions that can be executed by the subsystem and sent to the corresponding subsystem execution terminal. The running status feedback information generated after the control instruction sequence is executed is used to update the preset cooperative control constraints.

[0171] The optimal combination of control parameters is an abstract set of parameters that needs to be transformed into a sequence of control commands that each subsystem can understand and execute. This transformation process requires consideration of the communication protocols and command formats of each subsystem to ensure the accuracy and executability of the commands. After the control command sequence is sent, the subsystem execution end will operate according to the commands and return operational status feedback information. This feedback information will be used to optimize and update the preset cooperative control constraints, making the constraints more consistent with the actual operating conditions.

[0172] Step S151: Analyze the specific adjustment values ​​and execution order in the optimal adjustment parameter combination, classify the specific adjustment values ​​according to subsystems, match them with the operating parameter types of each subsystem, and determine the parameter name and adjustment method corresponding to each adjustment value.

[0173] The optimal combination of adjustment parameters is analyzed to extract the specific adjustment values ​​and their execution order. These specific adjustment values ​​are categorized according to the air conditioning system, lighting system, elevator system, and security system, and then matched with the operating parameter types of each subsystem. For example, the specific adjustment values ​​for the air conditioning system are matched with temperature parameters and air volume parameters, and the corresponding parameter names are determined as "return air temperature setpoint," "supply air volume setpoint," etc., with adjustment methods such as "incremental adjustment" and "absolute adjustment." Incremental adjustment increases or decreases the adjustment value based on the current parameter value, while absolute adjustment directly sets the parameter value to the adjustment value.

[0174] Step S152: Add timing markers to the adjustment instructions of each subsystem according to the execution order.

[0175] The execution order is determined based on the numerical values ​​of the timing parameters. Timing markers, such as "Timing 1," "Timing 2," etc., are added to the adjustment instructions of each subsystem according to this order. These timing markers instruct the subsystem execution end to execute the instructions in the correct order, ensuring coordinated operation between the subsystems. For example, if Timing 1 corresponds to the adjustment instructions for the security system and Timing 2 corresponds to the adjustment instructions for the lighting system, the subsystem execution end will execute the instructions for Timing 1 first, and then execute the instructions for Timing 2.

[0176] Step S153: Integrate the parameter name, adjustment value, adjustment mode and timing mark into a control instruction unit. Each subsystem corresponds to a set of control instruction units. All control instruction units are sorted according to the timing mark to form a control instruction sequence.

[0177] Each subsystem's parameter name, adjustment value, adjustment method, and timing mark together constitute a control instruction unit. For example, the control instruction unit of an air conditioning system includes the parameter name "return air temperature setpoint", the adjustment value "a certain value", the adjustment method "absolute adjustment", and the timing mark "timing 3".

[0178] The control command units of all subsystems are sorted according to their timing marks to form a complete control command sequence. The control command unit with the smaller timing mark is placed earlier, ensuring that the subsystem can execute adjustment operations in the predetermined order.

[0179] Step S154: Send the control command sequence to the corresponding subsystem execution terminal, so that after the subsystem execution terminal executes the control command sequence, it collects the running status parameters after execution and generates running status feedback information, which includes the adjusted parameter values, execution time and response characteristics.

[0180] Through the communication network between subsystems, a sequence of control commands is sent to the corresponding subsystem execution terminals, such as air conditioning controllers, lighting controllers, elevator control cabinets, and security control hosts. After receiving the command sequence, the subsystem execution terminal executes the corresponding operation according to the timing markers and adjustment methods.

[0181] After execution, the subsystem execution end collects the adjusted operating status parameters through sensors, such as the adjusted return air temperature of the air conditioning system and the adjusted brightness value of the lighting system, records the execution time and response characteristics, such as the time from the start of the instruction to its completion and the parameter fluctuations during the execution process, and generates operating status feedback information.

[0182] Step S155: Compare the running status feedback information with the historical data in the preset collaborative control constraints, calculate the deviation, and adjust the threshold range and benchmark features in the collaborative control constraints based on the deviation to complete the update of the constraints.

[0183] The adjusted parameter values, execution time, and response characteristics in the operational status feedback information are compared with historical data in the collaborative control constraints. The deviation between the two is calculated, such as the difference between the adjusted temperature value and the temperature value under the same operating conditions in the historical data, and the difference between the execution time and the historical average execution time.

[0184] Based on the magnitude and direction of the deviation, adjust the threshold range and baseline characteristics in the collaborative control constraints. For example, if multiple feedback messages show that the actual adjustment value of a certain parameter is generally higher than historical data and the operation is good, then appropriately increase the upper limit of the threshold for that parameter; if the response characteristics show that there is a delay in the coordination between subsystems, then adjust the response delay threshold in the associated baseline characteristics.

[0185] Through these continuous updates, the collaborative control constraints can adapt to the operational changes of various subsystems within the building, thereby improving the accuracy and effectiveness of multi-system collaborative control.

[0186] Figure 2 The illustration shows exemplary hardware and software components of a multi-system collaborative control system 100 for building intelligence, which can implement the ideas of this application, according to some embodiments of this application. For example, processor 120 can be used in the multi-system collaborative control system 100 for building intelligence and to perform the functions in this application.

[0187] For example, a multi-system collaborative control system 100 applied to building intelligence may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and various forms of storage media 140, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the multi-system collaborative control system 100 applied to building intelligence may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The methods of this application can be implemented according to these program instructions. The multi-system collaborative control system 100 applied to building intelligence also includes an I / O interface 150 between the computer and other input / output devices.

[0188] Furthermore, this embodiment of the invention also provides a readable storage medium, wherein computer-executable instructions are preset in the readable storage medium, and when the processor executes the computer-executable instructions, the above-mentioned multi-system collaborative control method applied to building intelligence is implemented.

[0189] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A multi-system collaborative control method applied to building intelligence, characterized in that, The method includes: Collect the operational status parameter streams of each subsystem within the building and generate a target status feature set, which includes the operational mode features, resource occupancy features, and interactive response features of each subsystem. The target state feature set is matched and analyzed with the preset collaborative control constraints to generate a collaborative correlation coefficient set between subsystems. The collaborative correlation coefficient set is used to characterize the degree of mutual influence between different subsystems during operation. A collaborative adjustment target model is constructed based on the collaborative correlation coefficient set and the target state feature set. The variables of the collaborative adjustment target model include the adjustment amount of the operating parameters and the adjustment time sequence parameters of each subsystem. Solving the collaborative adjustment target model yields the optimal combination of adjustment parameters that satisfies the constraints. The optimal combination of adjustment parameters includes the specific adjustment values ​​and execution order of each subsystem. The optimal combination of adjustment parameters is converted into a sequence of control instructions that can be executed by the subsystem and sent to the corresponding subsystem execution terminal. The operating status feedback information generated after the control instruction sequence is executed is used to update the preset cooperative control constraints.

2. The multi-system collaborative control method for building intelligence according to claim 1, characterized in that, The process involves collecting the operational status parameter streams of each subsystem within the building and generating a target status feature set, including: The collected operating status parameter streams of each subsystem are subjected to feature parsing to extract pattern recognition feature values ​​from the operating status parameter streams. The pattern recognition feature values ​​include parameter fluctuation period, parameter change trend and parameter correlation. Based on the pattern recognition feature values, an operation mode feature is generated. The operation mode feature includes a continuous operation mode identifier, an intermittent operation mode identifier, and an alternating operation mode identifier. Each mode identifier corresponds to a set of pattern recognition feature value ranges. Resource usage-related parameters are extracted from the running status parameter stream, and resource usage characteristics are calculated. The resource usage characteristics include resource allocation ratio, resource switching frequency, and resource idle time. The resource allocation ratio is determined based on the ratio of resource consumption parameters to total resource parameters in the parameter stream. The response parameters of each subsystem during the parameter interaction process are captured to generate interaction response features. The interaction response features include response delay features, response stability features and response correlation features. The response delay features are determined based on the interval parameter between parameter sending and receiving. The operating mode characteristics, resource consumption characteristics, and interactive response characteristics are classified and integrated according to subsystems to form a target state characteristic set. The characteristics of each subsystem correspond to independent feature items in the target state characteristic set.

3. The multi-system collaborative control method for building intelligence according to claim 2, characterized in that, The generation of operating mode features based on the pattern recognition feature values ​​includes: Define the intervals for pattern recognition feature values. These intervals are determined based on the statistical results of historical operating data of the building subsystem and include continuous pattern intervals, intermittent pattern intervals, and alternating pattern intervals. The extracted pattern recognition feature values ​​are compared with the defined intervals. When the parameter fluctuation period is in the continuous mode interval, the parameter change trend is linear, and the parameter correlation is in the corresponding interval, it is marked as a continuous operation mode identifier. When the parameter fluctuation cycle is in the intermittent mode range, the parameter change trend is step-like, and the parameter correlation is in the corresponding division range, it is marked as an intermittent operation mode identifier. When the parameter fluctuation cycle alternates between the continuous mode range and the intermittent mode range, the parameter change trend switches periodically, and the parameter correlation changes alternately with the cycle, it is marked as an alternating operation mode identifier. Each operating mode identifier is bound to a corresponding pattern recognition feature value range to form an operating mode feature that includes the identifier type and feature range. The feature range includes the interval of parameter fluctuation period, the feature description of parameter change trend, and the interval of parameter correlation.

4. The multi-system collaborative control method for building intelligence according to claim 1, characterized in that, The step of matching and analyzing the target state feature set with preset collaborative control constraints to generate a set of collaborative correlation coefficients between subsystems includes: The preset collaborative control constraints are analyzed, and the associated benchmark features and influence threshold ranges in the constraints are extracted. The associated benchmark features include subsystem type matching features, operating parameter compatibility features, and resource sharing features. Each subsystem feature in the target state feature set is compared with the associated benchmark feature item by item to calculate the feature matching degree, which is determined based on the overlap and similarity of feature items. The initial correlation coefficient is calculated based on the feature matching degree and the influence threshold range. When the feature matching degree is in the upper limit range of the influence threshold range, the initial correlation coefficient takes the corresponding high value. When the feature matching degree is in the lower limit range of the influence threshold range, the initial correlation coefficient takes the corresponding low value. The initial correlation coefficients are cross-validated. The validation process is based on the actual impact data of different subsystems in historical collaborative operation, and the initial correlation coefficients that deviate from the actual impact are adjusted. The verified and adjusted correlation coefficients are categorized and organized by subsystem to form a set of collaborative correlation coefficients. Each subsystem corresponds to a correlation value and a description of the correlation characteristics.

5. The multi-system collaborative control method for building intelligence according to claim 4, characterized in that, The analysis of the preset collaborative control constraints extracts the associated benchmark features and influence threshold ranges from the constraints, including: The preset collaborative control constraints are decomposed into condition items, feature items, and threshold items. The condition items include the preconditions for collaborative operation, the feature items include the associated benchmark features, and the threshold items include the range of influencing thresholds. Extract subsystem type matching features from the feature items. The subsystem type matching features include type compatibility identifiers, functional complementarity identifiers, and hierarchical correspondence identifiers. The type compatibility identifier is used to indicate whether different types of subsystems can operate collaboratively. Extracting operational parameter compatibility features, which include parameter range overlap, parameter change trend consistency, and parameter unit adaptability, wherein the parameter range overlap is determined based on the overlapping portion of the value ranges of the operational parameters of each subsystem; Extract resource sharing features, which include resource type matching degree, resource allocation conflict rate, and resource reuse efficiency. The resource type matching degree is determined based on the number of overlapping categories of resources required by the subsystem. The influence threshold range is extracted from the threshold items. The influence threshold range includes the lower limit of the correlation coefficient, the upper limit of the correlation coefficient, and the critical transformation value. Each correlation benchmark feature corresponds to a set of influence threshold ranges.

6. The multi-system collaborative control method for building intelligence according to claim 1, characterized in that, The construction of the collaborative adjustment target model based on the collaborative correlation coefficient set and the target state feature set includes: Each correlation coefficient in the set of collaborative correlation coefficients is used as a weighting factor and multiplied with the interaction response feature in the corresponding target state feature set to generate the correlation influence value between subsystems; Using the adjustment amount of the operating parameters of each subsystem as variables, and combining the associated impact value, the first part of the collaborative adjustment target model is constructed to characterize the sum of the associated impact after the adjustment of the subsystems; Extract resource occupancy features from the target state feature set, multiply the resource occupancy features by the adjustment time series parameters, and generate a feature quantity of resource consumption changing over time, which serves as the second part of the collaborative adjustment target model; The constraints of the coordinated adjustment target model are set, including the range of values ​​of the operation parameter adjustment amount, the sequential relationship constraints of the adjustment time parameters, and the upper limit constraints of the associated influence values. Each constraint corresponds to a set of boundary conditions determined based on historical data. The first part, the second part, and the constraint terms are integrated into a collaborative adjustment target model. The output value of the collaborative adjustment target model is the comprehensive adjustment benefit value, which corresponds to the sum of related impacts and resource consumption characteristics.

7. The multi-system collaborative control method for building intelligence according to claim 6, characterized in that, The constraints for setting the collaborative adjustment target model include: Extract the parameter fluctuation range of each subsystem under safe operating conditions from historical operating data, and use the extreme value of the parameter fluctuation range as the boundary of the value range of the operating parameter adjustment amount to form the first constraint sub-item; By analyzing the correlation data between the adjustment execution order and adjustment effect of each subsystem in the historical coordinated adjustment process, the rules for the order of adjustment timing parameters are determined. When the adjustment result of one subsystem affects the adjustment basis of another subsystem, the adjustment timing parameters of the first subsystem must be earlier than the adjustment timing parameters of the second subsystem, thus forming the second constraint item. The maximum safe threshold for the associated impact value of each subsystem in the historical operation is calculated. When the associated impact value exceeds the maximum safe threshold, it will cause an operational abnormality. The maximum safe threshold is used as the upper limit constraint of the associated impact value, forming the third constraint sub-item. An adjustment coefficient for the boundary conditions is set for each constraint sub-item. The adjustment coefficient is determined based on the operating mode features in the current target state feature set. The adjustment coefficient for the continuous operating mode is smaller than the adjustment coefficient for the intermittent operating mode. The first constraint sub-item, the second constraint sub-item, the third constraint sub-item, and the corresponding adjustment coefficient are integrated into the constraint terms of the collaborative adjustment target model, and the constraint terms are incorporated into the collaborative adjustment target model in the form of inequalities.

8. The multi-system collaborative control method for building intelligence according to claim 1, characterized in that, Solving the collaborative adjustment target model to obtain the optimal combination of adjustment parameters that satisfies the constraints includes: The collaborative adjustment target model is solved using a feature space search method, with the adjustment amount of the running parameters and the adjustment time sequence parameters as the dimensions of the search space, and the value range of each dimension corresponding to the boundary conditions in the constraint terms. Within the search space, an initial parameter combination sample is generated. The number of the initial parameter combination samples is determined based on the number of subsystems and the parameter dimension. Each initial parameter combination sample contains a set of operating parameter adjustment amounts and adjustment timing parameters. Calculate the output value of the collaborative adjustment target model corresponding to each initial parameter combination sample, and select the samples whose output values ​​meet the preset optimization target as candidate samples; The candidate samples are iteratively optimized by adjusting the values ​​of the running parameters and the order of the time-series parameters, so that the output value of the collaborative adjustment target model changes in a better direction. All conditions of the constraint terms must be met during the iteration process. When the output value of the coordinated adjustment target model after iterative optimization stabilizes within a preset range and no longer changes, the iteration stops, and the corresponding parameter combination is determined as the optimal adjustment parameter combination. The specific adjustment value in the optimal adjustment parameter combination retains the parameter dimension corresponding to the target state feature set, and the execution order is determined based on the numerical value of the adjustment timing parameter.

9. The multi-system collaborative control method for building intelligence according to claim 1, characterized in that, The step of converting the optimal adjustment parameter combination into a sequence of control instructions executable by the subsystem and sending it to the corresponding subsystem execution terminal includes: The specific adjustment values ​​and execution order in the optimal adjustment parameter combination are analyzed. The specific adjustment values ​​are classified by subsystem and matched with the operating parameter types of each subsystem to determine the parameter name and adjustment method corresponding to each adjustment value. Add timing markers to the adjustment instructions of each subsystem according to the order of execution; The parameter name, adjustment value, adjustment method and timing mark are integrated into a control instruction unit. Each subsystem corresponds to a set of control instruction units. All control instruction units are sorted according to timing mark to form a control instruction sequence. The control command sequence is sent to the corresponding subsystem execution terminal so that after the subsystem execution terminal executes the control command sequence, it collects the running status parameters after execution and generates running status feedback information, which includes the adjusted parameter values, execution time and response characteristics. The operational status feedback information is compared with historical data in the preset collaborative control constraints to calculate the deviation. Based on the deviation, the threshold range and benchmark features in the collaborative control constraints are adjusted to complete the constraint update.

10. A multi-system collaborative control system for building intelligence, characterized in that, The system includes a processor and a memory, the memory being connected to the processor. The memory is used to store programs, instructions, or code, and the processor is used to execute the programs, instructions, or code in the memory to implement the multi-system collaborative control method for building intelligence as described in any one of claims 1-9.

Images