Control method for central air conditioner terminal detection device

Abstract

The invention relates to the technical field of device control, and provides a control method for a central air conditioner tail end detection device, which is characterized in that scene dimension parameters are comprehensively analyzed, dynamic reference values of basic operation parameters are accurately determined, and quick identification and positioning of abnormity are realized in combination with multi-dimension verification; according to the method, the limitation that traditional central air conditioner control depends on fixed parameter setting and single abnormity judgment is broken through, the self-adaptive capacity of a detection device to complex scenes is remarkably improved, the operation stability of the central air conditioner is improved, and the method is suitable for popularization and application. The abnormal response time is greatly shortened, the energy consumption is reduced, the problems that an existing central air conditioner is low in operation efficiency, inaccurate in abnormal diagnosis, waste in energy and the like are effectively solved, more comfortable, energy-saving and intelligent use experience is provided for a user, meanwhile, the operation and maintenance cost is reduced, and the service life of the central air conditioner is prolonged.

Description

Technical Field

[0001] The present application relates to the field of device control technology, and in particular to a control method for a central air-conditioning terminal detection device. Background Art

[0002] In the field of modern buildings, central air conditioning is becoming increasingly important as a key facility to ensure indoor environmental comfort and stable equipment operation. With the continuous growth of demand for building intelligence and energy conservation, higher requirements are placed on the control accuracy, abnormality diagnosis efficiency and energy utilization efficiency of central air conditioning terminal detection equipment. Currently, existing technologies mainly use fixed parameter settings and simple threshold judgment methods in central air conditioning terminal control.

[0003] From the current development status, early central air-conditioning control systems relied on manual experience to set the baseline values ​​of basic operating parameters. These baseline values ​​usually remained fixed during the operation of the central air-conditioning, and were difficult to adapt to the differences in operating requirements brought about by different areas, different usage scenarios and seasonal changes in the building. With the development of technology, some control systems introduced simple sensor feedback mechanisms, which can make preliminary abnormal judgments and equipment adjustments based on the comparison of single parameters such as temperature and flow with preset thresholds. However, this method can only handle simple abnormal situations and cannot deeply analyze the causes of complex abnormalities.

[0004] Existing technologies suffer from numerous shortcomings, including fixed baseline settings that result in poor adaptability of detection devices; a single-dimensional anomaly diagnosis system that relies too heavily on a single parameter threshold, making it inaccurately identifying complex anomaly types and locating their sources; and a lack of intelligent response and self-optimization capabilities. Most technologies fail to address the problem of how to identify anomaly types and locate their sources in complex scenarios through parameter comparison and multi-dimensional anomaly verification, enabling closed-loop optimization. Summary of the Invention

[0005] In response to the shortcomings of the prior art, the present application provides a control method for a central air-conditioning terminal detection device, the method comprising: identifying whether the regional temperature of the central air-conditioning terminal is abnormal, and verifying whether the basic operating parameters of the central air-conditioning terminal deviate when the regional temperature is abnormal;

[0006] The basic operating parameters and scenario dimension parameters of the central air conditioning terminal are obtained in real time. Based on the scenario dimension parameters, the baseline value offset of the basic operating parameters is generated through the compensation rule library. The baseline value offset is superimposed through the time series prediction method to determine the dynamic baseline value and dynamic threshold range;

[0007] Compare basic operating parameters with dynamic benchmark values. In response to deviations of basic operating parameters from the dynamic benchmark values, set an anomaly detection method to fuse the associated features of scenario dimension parameters and basic operating parameters to form a fused feature set. Based on the fused feature set, verify the distribution properties, local performance, and scenario requirements of the basic operating parameters to identify the anomaly type and locate the anomaly source.

[0008] Extract the attribute characteristics of the abnormal type and abnormal source, match the scenario response strategy and control the central air-conditioning terminal detection device to adjust the central air-conditioning terminal, while monitoring the changing trend of basic operating parameters to update the compensation rule base.

[0009] As an optional implementation manner, the dynamic reference value determination logic includes:

[0010] The basic operating parameters of central air conditioning terminals are predicted at multiple scales, including short-term, medium-term and long-term scales, using the time series prediction method.

[0011] Fusion weights are assigned according to the time length of the scales. The prediction results of different scales are fused based on the fusion weights to obtain the initial benchmark value. The consistency of the prediction results of different scales is verified, and the prediction results of different scales are corrected by combining historical adjustment data and scene dimension parameters.

[0012] The baseline value offset and the initial baseline value are superimposed to determine the dynamic baseline value, and the confidence interval of the prediction result is determined based on the error of the historical prediction result to generate the dynamic threshold interval.

[0013] As an optional implementation manner, the reference value offset generation sub-logic includes:

[0014] Obtain scene dimension parameters of central air conditioning terminals in real time, pre-process and analyze them to obtain thermal height coefficient, heat flow direction identifier, load adjustment factor and topology influence matrix;

[0015] Extract the compensation rule base, determine the vertical thermal compensation amount based on the thermal height coefficient and heat flow direction identifier, determine the occasion load compensation amount based on the load adjustment factor and air volume and water flow, and monitor the valve opening of the bypass pipe to determine the terminal impact compensation amount based on the topological impact matrix and the temperature change of the adjacent terminal;

[0016] The vertical thermal compensation, occasion load compensation and terminal impact compensation are processed by weighted summation to generate a reference value offset and recorded in the compensation rule base.

[0017] As an optional implementation, the parsing sub-logic of the scene dimension parameter includes:

[0018] Obtain the vertical height of the central air conditioning terminal in real time, determine the location area of ​​the central air conditioning terminal based on the building structure, and assign a thermal height coefficient to the location area based on the principles of building thermodynamics;

[0019] Obtain the seasonal operation mode and occasion type identification of the central air-conditioning terminal in real time, encode the seasonal operation mode into the heat flow direction identification, and map the occasion type identification into the load adjustment factor;

[0020] The bypass pipe status and adjacent terminal relationship of the central air-conditioning terminal are obtained in real time, and a topological influence matrix is ​​constructed with the central air-conditioning terminal as the matrix node. The matrix elements of the topological influence matrix represent the hydraulic coupling strength between the central air-conditioning terminals.

[0021] As an optional implementation, the abnormality source location logic includes:

[0022] Calculate the outliers for distribution anomalies, local performance anomalies, and scenario demand anomalies respectively, and dynamically assign verification weights for distribution anomalies, local performance anomalies, and scenario demand anomalies based on the scenario type to fuse them to obtain the total value of dimensional anomalies.

[0023] Based on the hydraulic topology, the node orientation is determined based on the total value of dimensional anomalies to form the direction of anomaly propagation. The hydraulic coupling strength is also used to determine the probability of anomaly propagation on each edge. Furthermore, the central air conditioning terminals with anomalies are density clustered based on the total value of dimensional anomalies to identify anomaly clusters.

[0024] If the abnormal cluster includes a single central air-conditioning terminal, the abnormal source is determined to be a terminal monomer and the attribute characteristic of the abnormal source is local failure; if the abnormal cluster includes multiple central air-conditioning terminals, and it is judged that the edges with high probability of abnormal propagation are densely distributed in the abnormal cluster, the abnormal source is determined to be a group and the attribute characteristic of the abnormal source is cascade risk.

[0025] As an optional implementation, the verification sub-logic of the distribution property includes:

[0026] Obtain the vertical position of the central air conditioning terminal in real time, and divide different floors into thermal characteristic areas based on the building structure and the location area of ​​the central air conditioning terminal;

[0027] Extract the heat flow direction identifier encoded according to the seasonal operation mode, and determine the spatiotemporal constraints of the basic operation parameter distribution based on the thermal characteristic area;

[0028] The basic operating parameters of central air-conditioning terminals on the same floor are density clustered, outlier terminals that deviate from the cluster center are automatically identified, and the residuals between the basic operating parameters of the outlier terminals and the spatiotemporal constraints are calculated. When the residuals are not within the dynamic threshold range, it is determined that there is an abnormal distribution property.

[0029] As an optional implementation, the local performance verification sub-logic includes:

[0030] The topological influence matrix constructed based on the bypass pipe status and the relationship between adjacent terminals is converted into a hydraulic topology graph. The central air conditioning terminals are used as nodes. The connection relationship between the central air conditioning terminals represents the edge of the hydraulic topology graph. The edge weight represents the hydraulic coupling strength in the topological influence matrix. The valve opening of the bypass pipe is obtained in real time to dynamically update the edge weight of the hydraulic topology graph.

[0031] When it is detected that the basic operating parameters of the central air-conditioning terminal deviate from the dynamic reference value, the conduction path between the central air-conditioning terminal and the adjacent terminal is determined by the edges and edge weights of the hydraulic topology graph;

[0032] The deviation value is obtained by subtracting the theoretical parameter value of the adjacent ends in the conduction path from the actual basic operating parameter. When the deviation value is not within the dynamic threshold range, it is determined that there is a local performance abnormality.

[0033] As an optional implementation, the verification sub-logic of the scenario requirement includes:

[0034] Obtain the occasion type identification of the central air-conditioning terminal in real time, parse the occasion type identification and extract the basic operating parameters under the occasion type identification according to the scene demand database to form a scene demand set;

[0035] The basic operating parameters obtained in real time are matched with the scenario requirement set. If the basic operating parameters do not match the scenario requirement set, it is determined that there is a scenario requirement anomaly.

[0036] Continuously record the verification results and basic operating parameters of each scenario requirement. When there are multiple abnormal scenario requirements, analyze the historical basic operating parameters to optimize the scenario requirement database.

[0037] As an optional implementation, the matching logic of the scenario response strategy includes:

[0038] Establish scenario response strategies, classify them into three levels according to the anomaly type, anomaly source, and attribute characteristics of the anomaly source, and set an index identifier for each scenario response strategy;

[0039] Obtain the identification results of the attribute characteristics of the anomaly type and anomaly source in real time, and match them in the scenario response strategy according to the three-level classification. If multiple scenario response strategies are matched, the priority is determined according to the scenario type identifier;

[0040] Execute the matching scenario response strategy and monitor the changing trend of each basic operating parameter in the central air-conditioning terminal in real time to provide feedback and modify the scenario response strategy.

[0041] As an optional implementation, the update logic of the compensation rule base includes:

[0042] Real-time monitoring of the deviation of the basic operating parameters of the central air-conditioning terminal from the dynamic reference value, including the deviation amplitude and degree, and judging whether to update the compensation rule base according to the deviation;

[0043] Analyze the correlation between scenario dimension parameters and deviations from basic operating parameters to generate candidate rules, simulate the verification effect of executing candidate rules when basic operating parameters deviate from dynamic baseline values, and select compensation rules based on the verification effect;

[0044] The filtered compensation rules are added to the compensation rule library, and priorities are assigned according to the applicable frequency of the compensation rules. The changing trend of each basic operating parameter in the central air-conditioning terminal is monitored in real time to iteratively update the compensation rule library.

[0045] Compared with the existing technology, the beneficial effects of the present application are: through comprehensive analysis of scene dimension parameters, the dynamic baseline values ​​of basic operating parameters are accurately determined, and multi-dimensional verification is combined to achieve rapid identification and positioning of anomalies, and then the scene response strategy is matched according to the attribute characteristics of the anomaly type and the source of the anomaly and the compensation rule base is continuously optimized. This method breaks the limitations of traditional central air-conditioning control that relies on fixed parameter settings and single anomaly judgment, significantly improves the detection device's adaptability to complex scenarios, improves the stability of central air-conditioning operation, greatly shortens the anomaly response time, and reduces energy consumption. It effectively solves the problems of low operating efficiency, inaccurate anomaly diagnosis and energy waste in existing central air-conditioning, and provides users with a more comfortable, energy-saving and intelligent use experience, while reducing operation and maintenance costs and extending the service life of central air-conditioning.

[0046] Based on multi-dimensional scenario parameters such as the terminal vertical height and occasion type identification, the benchmark value offset is generated through the compensation rule library, and the dynamic benchmark value and dynamic threshold range are determined in combination with the time series prediction method. The thermal characteristics of the building, the usage scenario requirements and the time change law are fully considered, so that the dynamic benchmark value can reflect the reasonable parameter range under the current operating scenario in real time, thereby ensuring the precise operation of the central air-conditioning, avoiding problems such as temperature discomfort and energy waste caused by unreasonable benchmark values, and improving the accuracy and energy efficiency of the central air-conditioning operation.

[0047] When basic operating parameters deviate from dynamic benchmark values, they are verified from three dimensions: distribution properties, local performance, and scenario requirements. By verifying the distribution properties based on the vertical position of the terminal and the seasonal operation mode, anomalies caused by abnormal thermal distribution of the building can be discovered; local performance is verified by the bypass pipe status and the relationship between adjacent terminals, which can quickly locate hydraulic system-related anomalies; scenario requirements are verified based on the occasion type identification, which can identify parameter setting problems that do not meet the scenario operation requirements; comprehensive analysis of multi-dimensional verification results can accurately identify the type of anomaly and locate the source of the anomaly. Compared with traditional anomaly diagnosis methods, this greatly improves the accuracy and comprehensiveness of anomaly diagnosis, reduces misjudgments and missed judgments, shortens anomaly investigation time, and provides strong support for quickly repairing anomalies and ensuring stable operation of central air conditioning.

[0048] The scenario response strategy is matched based on the attribute characteristics of the anomaly type and the source of the anomaly to ensure that the most appropriate processing plan can be quickly called when an anomaly occurs, thereby improving the pertinence and effectiveness of anomaly processing. At the same time, during the execution of the scenario response strategy, the changes in basic operating parameters are monitored to update the compensation rule library to make the compensation rules more in line with actual operating needs. This closed-loop optimization mechanism realizes self-learning and adaptive adjustment of the detection device. As the operating data accumulates, the detection device's adaptability to complex scenarios continues to increase, and its operating efficiency continues to improve, further reducing energy consumption and operation and maintenance costs, and ensuring the long-term stable and efficient operation of central air conditioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be derived from these drawings without inventive work. Among them:

[0050] Figure 1 A flow chart of a control method for a central air-conditioning terminal detection device provided in an embodiment of the present application;

[0051] Figure 2 This is a sub-logic diagram of the scene dimension parameters of a control method for a central air-conditioning terminal detection device provided in an embodiment of the present application;

[0052] Figure 3 A sub-logic diagram for verifying the distributed nature of a control method for a central air-conditioning terminal detection device provided in an embodiment of the present application;

[0053] Figure 4 A sub-logic diagram for verifying the local performance of a control method for a central air-conditioning terminal detection device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] In order to make the objectives, technical solutions and advantages of the embodiments of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application are clearly and completely described below in conjunction with the drawings in the specification. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0055] like Figure 1 As shown, a method flow chart of a control method for a central air-conditioning terminal detection device is provided in an embodiment of the present application, and the method includes:

[0056] First, identify whether the regional temperature of the central air-conditioning terminal is abnormal. When the regional temperature is abnormal, verify whether the basic operating parameters of the central air-conditioning terminal are deviated, so as to subsequently identify the abnormal type and locate the abnormal source.

[0057] S1. Obtain the basic operating parameters and scenario dimension parameters of the central air-conditioning terminal in real time, generate the baseline value offset of the basic operating parameters based on the scenario dimension parameters through the compensation rule library, and superimpose the baseline value offset through the time series prediction method to determine the dynamic baseline value and dynamic threshold range.

[0058] Basic operating parameters include air volume, water flow, valve opening and temperature.

[0059] Furthermore, the sub-logic for generating the reference value offset includes:

[0060] Obtain scene dimension parameters of central air conditioning terminals in real time, pre-process and analyze them to obtain thermal height coefficient, heat flow direction identifier, load adjustment factor and topology influence matrix;

[0061] Extract the compensation rule base, determine the vertical thermal compensation amount based on the thermal height coefficient and heat flow direction identifier, determine the occasion load compensation amount based on the load adjustment factor and air volume and water flow, and monitor the valve opening of the bypass pipe to determine the terminal impact compensation amount based on the topological impact matrix and the temperature change of the adjacent terminal;

[0062] The vertical thermal compensation, occasion load compensation and terminal impact compensation are processed by weighted summation to generate a reference value offset and recorded in the compensation rule base.

[0063] The operation of central air conditioning is affected by many scene factors. The original scene dimension parameters cannot be used directly for calculation. They need to be converted into quantifiable and standardized parameters to provide effective input for subsequent compensation calculation. Through the sensor installed on the central air conditioning terminal, the vertical height of the terminal is obtained in real time, and combined with the floor information of the building structure, it is judged whether the central air conditioning terminal is in the above-ground area or the underground area. Based on the principle of building thermodynamics, corresponding thermal height coefficients are assigned to different location areas. In specific applications, the underground area is assigned a lower thermal height coefficient due to the sinking of cold air; the high-rise area is assigned a higher thermal height coefficient due to the accumulation of hot air; for the seasonal operation mode, the heating or cooling mode is encoded as a heat flow direction identifier, where the heating mode is +1 and the cooling mode is -1; and the occasion type identifier is matched with the preset occasion load feature library through semantic recognition technology to be converted into a load adjustment factor, where a higher load adjustment factor corresponds to shopping mall promotions.

[0064] The bypass pipe status is monitored in real time by sensors installed on the bypass pipe for its open and closed status and valve opening. The relationship between adjacent terminals is obtained through the data communication network between the central air-conditioning terminals to obtain connection information, and then a topological influence matrix is ​​constructed. The matrix elements are taken according to the hydraulic coupling between the terminals, where the value is 1 if there is coupling and 0 otherwise. This achieves standardized processing of scene dimension parameters, converts complex scene information into quantitative indicators that can be identified and calculated by the detection device, enables the detection device to accurately perceive the characteristics of the current operating scene, and lays the foundation for the subsequent accurate calculation of the compensation amount.

[0065] Different scenario dimension parameters have different effects on the basic operating parameters of the central air-conditioning terminal. It is necessary to calculate each compensation amount separately to correct the baseline value of the basic operating parameter to make it meet the actual needs of the current operating scenario; combined with the corresponding logic in the compensation rule library, the vertical thermal compensation amount is determined according to the thermal height coefficient and the heat flow direction identifier. Preferably, in the cooling mode, the thermal height coefficient of the high-rise area is higher, and the corresponding cooling capacity compensation is increased according to the compensation rule; and for the occasion load compensation amount, the load adjustment factor is combined with the real-time monitored air volume and water flow, and calculated according to the calculation formula in the compensation rule library, where the occasion load compensation amount = (load adjustment factor-1) × basic benchmark value , and the basic benchmark value = water flow / air volume is obtained to meet the load requirements in different occasions. At the same time, the valve opening of the bypass pipe is continuously monitored. According to the topological influence matrix and the temperature changes of the adjacent terminals, the terminal impact compensation amount is calculated according to the relevant rules in the compensation rule library. The terminal impact compensation amount is the sum of the products of the hydraulic coupling intensity in the topological influence matrix and the temperature changes of the adjacent terminals, and the hydraulic and thermal interactions between adjacent terminals need to be considered; thus, targeted calculations are made on the impact of different scenario factors on the basic operating parameters, and the benchmark values ​​can be accurately adjusted to make the basic operating parameters of the central air-conditioning terminal more in line with the actual scenario requirements, thereby improving the energy efficiency and comfort of the detection device operation.

[0066] The various compensation quantities are integrated to obtain a unified benchmark value offset, which is used to adjust the benchmark values ​​of the basic operating parameters. At the same time, the calculation process and results are recorded in the compensation rule library to provide actual operation data support for subsequent rule optimization; the weight distribution strategy of weighted summation is obtained from the compensation rule library, and the vertical thermal compensation quantity, occasion load compensation quantity and terminal impact compensation quantity are weightedly summed according to the distribution strategy to obtain the benchmark value offset; then the scene dimension parameters involved in this calculation, the calculation process of each compensation quantity and the final offset result are recorded in the compensation rule library according to specific data format and storage rules; thereby accurately calculating the benchmark value offset, providing key parameters for the determination of the dynamic benchmark value, and recording the calculation process and results in the compensation rule library, which helps to continuously optimize the compensation rules according to actual operating conditions, improve the adaptability of the detection device to different scenarios and the accuracy of long-term operation, and the generated benchmark value offset will be superimposed with the initial benchmark value determined subsequently to obtain the dynamic benchmark value, which is necessary data in the process of determining the dynamic benchmark value.

[0067] Among them, when the compensation rule library is initially constructed, based on the principles of building thermodynamics, fluid mechanics and the operating characteristics of central air conditioning, the corresponding determination methods for vertical thermal compensation, occasion load compensation and terminal impact compensation are preset to be solidified in the compensation rule library, and a large number of industry standard parameters and historical operating data are imported for construction.

[0068] Furthermore, if Figure 2 As shown, the parsing sub-logic of the scene dimension parameters includes:

[0069] Obtain the vertical height of the central air conditioning terminal in real time, and determine the location area of ​​the central air conditioning terminal based on the building structure. The location area includes the above-ground area and the underground area. Based on the principles of building thermodynamics, the location area is assigned a thermal height coefficient;

[0070] Obtain the seasonal operation mode and occasion type identification of the central air-conditioning terminal in real time, encode the seasonal operation mode into the heat flow direction identification, and map the occasion type identification into the load adjustment factor;

[0071] The bypass pipe status and adjacent terminal relationship of the central air-conditioning terminal are obtained in real time, and a topological influence matrix is ​​constructed with the central air-conditioning terminal as the matrix node. The matrix elements of the topological influence matrix represent the hydraulic coupling strength between the central air-conditioning terminals.

[0072] The central air-conditioning terminal has different vertical positions in the building, and the thermal impact it is subjected to varies. Converting the terminal vertical height into a thermal height coefficient can quantify this impact and provide a basis for subsequent parameter adjustments based on thermal characteristics. The terminal vertical height is obtained in real time and integrated with the obtained terminal vertical height and floor information of the building structure to determine whether the terminal is located in the above-ground area or the underground area. Based on the principles of building thermodynamics, the underground area is assigned a lower thermal height coefficient due to the sinking of cold air, while the above-ground area is assigned corresponding thermal height coefficients based on the rising hot air and heat accumulation at different heights. The determination of these thermal height coefficients is to accurately reflect the impact of the terminal vertical position on thermal-related parameters in subsequent calculations. Converting the terminal vertical height into a quantified thermal height coefficient enables the detection device to intuitively understand the differences in thermal characteristics of different terminal vertical positions, providing an accurate and quantitative basis for subsequent parameter adjustments based on thermal characteristics, and improving the adaptability of the detection device to operating scenarios with different terminal vertical positions and the accuracy of parameter adjustment.

[0073] Seasonal changes and different usage scenarios have very different operating requirements for central air-conditioning terminals. Converting seasonal operation modes and scenario type identifiers can enable the detection device to quickly identify the demand characteristics of the current operation scenario, thereby adjusting the operating parameters to meet actual needs; receiving seasonal operation mode instructions in real time, including heating mode or cooling mode, and encoding the heating mode as a heat flow direction identifier +1 and the cooling mode as -1 through encoding technology; and for the scenario type identifier, the input scenario type information is analyzed and processed through a semantic recognition algorithm. The scenario types include offices, shopping malls, hotels, etc., and the scenario type is matched with the pre-established scenario load feature library to obtain the scenario type information from the scenario. The corresponding load adjustment factor is extracted from the combined load feature database. The database stores load-related information such as personnel density and equipment usage in different occasions and different time periods. Through matching analysis, the load adjustment factor suitable for the current occasion is obtained, where the greater the load, the greater the load adjustment factor. The seasonal operation mode and occasion type identification are converted into quantitative indicators that can be used for calculation and control, so that the detection device can quickly identify the demand characteristics of the current operation scenario, and provide strong support for the precise adjustment of basic operation parameters. This improves the intelligence level of the detection device, can better meet the user's needs for central air-conditioning terminal operation in different scenarios, and improves the user experience.

[0074] The bypass pipe status and the relationship between adjacent terminals will affect the hydraulic conditions and heat transfer of the central air conditioning terminal. Constructing a topological influence matrix can quantify this influence relationship, providing a basis for analyzing terminal performance, adjusting operating parameters, and diagnosing abnormalities. Through the valve status sensor and opening sensor installed on the bypass pipe, the opening and closing status of the bypass pipe and the valve opening are obtained in real time, and the connection relationship between each central air conditioning terminal is also obtained. The central air conditioning terminal is used as a matrix node, and a topological influence matrix is ​​constructed based on the bypass pipe status and the adjacent terminal connection relationship. During the matrix construction process, if there is a hydraulic coupling relationship between central air conditioning terminal A and central air conditioning terminal B, that is, the bypass pipe is open and the two are connected in the hydraulic circuit, then the matrix element is assigned a value of 1, that is, the hydraulic coupling strength = 1. If there is no coupling relationship, that is, the bypass pipe is closed or physically unconnected, then the matrix element is assigned a value of 0, that is, the hydraulic coupling strength = 0. In this way, the complex connection relationship of central air conditioning terminals is converted into a matrix form that can be calculated and analyzed.

[0075] The topological influence matrix clearly presents the hydraulic connection relationship and mutual influence degree between central air-conditioning terminals, providing an intuitive and quantitative tool for the detection device to deeply analyze the local performance of the terminal, accurately determine the abnormal propagation path, and reasonably adjust the operating parameters. This helps to improve the abnormality diagnosis capability and operation optimization level of the detection device. In the subsequent local performance verification and abnormality source location process, the topological influence matrix plays a key role, directly affecting the accuracy and effectiveness of the operation control and abnormality handling process of the entire detection device.

[0076] Specifically, the dynamic reference value determination logic includes:

[0077] The basic operating parameters of central air conditioning terminals are predicted at multiple scales, including short-term, medium-term and long-term scales, using the time series prediction method.

[0078] Fusion weights are assigned according to the time length of the scales. The prediction results of different scales are fused based on the fusion weights to obtain the initial benchmark value. The consistency of the prediction results of different scales is verified, and the prediction results of different scales are corrected by combining historical adjustment data and scene dimension parameters.

[0079] The baseline value offset and the initial baseline value are superimposed to determine the dynamic baseline value, and the confidence interval of the prediction result is determined based on the error of the historical prediction result to generate the dynamic threshold interval.

[0080] The basic operating parameters of central air-conditioning terminals have different change patterns and fluctuation characteristics at different time scales. Single-scale predictions cannot fully reflect parameter changes. Multi-scale predictions can capture parameter change characteristics from multiple dimensions such as short-term fluctuations, medium-term trends, and long-term changes, providing more comprehensive and rich information for accurately determining dynamic benchmark values. Multi-scale predictions of basic operating parameters are performed through a variety of prediction algorithms. For short-term scales, the moving average algorithm is used to process the short-term scale. The moving average algorithm calculates the average value of the basic operating parameters in a recent period of time to predict the change trend of the basic operating parameters in the short term in the future, and can quickly respond to short-term fluctuations in the basic operating parameters. For medium-term scales, the long short-term memory network is used. The long short-term memory network can learn and memorize the basic operating parameters in the past few hours. By analyzing the change pattern within time, the load change situation at the medium-term scale is predicted by analyzing the historical basic operating parameters, and the complex change trend of the basic operating parameters at the medium-term scale is adapted to; for the long-term scale, the time series analysis method is used, according to the change law of the daily and weekly cycles in the historical basic operating parameters, while considering the impact of seasonal factors on the parameters, the long-term basic operating parameter trend is predicted, which can effectively cope with the seasonal and cyclical changes of the basic operating parameters; multi-scale prediction comprehensively analyzes the changes of the basic operating parameters from different time dimensions, makes full use of the change law of the basic operating parameters at different time scales, improves the accuracy and reliability of the prediction, enables the detection device to predict the parameter fluctuations at different time scales in advance, provides strong support for the reasonable adjustment of the basic operating parameters, and better meets user needs.

[0081] The prediction results of different scales reflect the changing trends of basic operating parameters in different time dimensions, but each has its own limitations. Through fusion, we can integrate information from various scales, make up for the shortcomings of single-scale predictions, and obtain more accurate initial benchmark values. At the same time, we can verify the consistency of the prediction results and make corrections, which can further improve the reliability of dynamic benchmark values ​​and make them more consistent with actual operating conditions. According to the length of the prediction scale, fusion weights are assigned to the prediction results of each scale. Generally speaking, short-term prediction results are closer to real-time conditions and reflect the current operating status more directly, so they are given higher weights. Although long-term prediction results are instructive for the overall trend, they are not The certainty is relatively large, and a relatively low weight is assigned; based on the assigned weights, the prediction results of different scales are weightedly fused to calculate the initial benchmark value. During the fusion process, the prediction results of different scales are verified for consistency, and whether there are conflicts or unreasonable differences between the prediction results of each scale. If inconsistencies are found, the historical adjustment data and the current scenario dimension parameters are combined for analysis. If the short-term prediction results and the medium-term prediction results are significantly different, and the current period is a shopping mall promotion, the basic operating parameter change data during the historical promotion period is referred to to revise the prediction results to make the prediction results of each scale more coordinated and consistent.

[0082] The fusion of prediction results fully integrates the prediction information of different time scales, improves the accuracy and reliability of the initial benchmark value, and the consistency verification and correction process effectively eliminates unreasonable prediction results, so that the initial benchmark value can more accurately reflect the actual operation situation, provide a more reliable reference basis for the stable operation of the detection device, and enhance the adaptability of the detection device to changes in basic operating parameters. The initial benchmark value obtained after fusion and correction will be superimposed with the benchmark value offset to determine the dynamic benchmark value. The accuracy of the initial benchmark value directly affects the calculation result of the dynamic benchmark value. It is a key intermediate result in the process of determining the dynamic benchmark value, and plays an important role in the subsequent system's judgment of whether the basic operating parameters are normal.

[0083] By superimposing the baseline value offset with the initial baseline value, a dynamic baseline value that comprehensively considers the current scenario and operating status can be obtained, providing an accurate reference standard for subsequent judgment of whether the basic operating parameters are normal. At the same time, a dynamic threshold interval is generated, which helps to more accurately judge parameter deviations and improve the accuracy of anomaly detection; the generated baseline value offset and the initial baseline value that has been fused and corrected are superimposed to obtain a dynamic baseline value, which can reflect the reasonable benchmark of the basic operating parameters under the influence of the current scenario dimension parameters. Then, based on the error analysis of historical prediction results, the confidence interval of the prediction results is determined. By statistically analyzing the deviation between the predicted value and the actual value in the historical prediction results, the mean and standard deviation of the error are calculated, and the appropriate confidence interval is determined based on statistical principles. Finally, with the dynamic baseline value as the center and combined with the confidence interval, a dynamic threshold interval is generated. When the basic operating parameters obtained in real time exceed the dynamic threshold interval, it will be judged that the basic operating parameters have deviated.

[0084] The determined dynamic benchmark value and dynamic threshold interval can be adjusted in real time according to the current scenario and operating status, providing a scientific and reasonable reference basis for accurately judging whether the basic operating parameters are normal. Compared with fixed benchmark values ​​and threshold intervals, the dynamic adjustment method can better adapt to various changes in the operation process of central air conditioning, improve the accuracy and timeliness of abnormal detection of detection devices, help ensure stable and efficient operation of central air conditioning, reduce the probability of abnormal occurrence and energy waste, dynamic benchmark value and dynamic threshold interval are important bases for judging whether the basic operating parameters are deviated, verifying the nature of parameters, and identifying abnormal types and locating abnormal sources. Accurate dynamic benchmark value and dynamic threshold interval can improve the accuracy of abnormal detection and diagnosis, and directly affect the effectiveness of subsequent abnormal handling processes.

[0085] From the above content, it can be seen that the dynamic benchmark value = initial benchmark value + benchmark value offset, where the initial benchmark value is obtained by weighted summation based on the short-term prediction results, the medium-term prediction results and the long-term prediction results. The dynamic benchmark value is updated according to the basic operating parameters obtained or when the scene dimension parameters change. When updating according to the basic operating parameters obtained, it is necessary to observe the update time of the short-term prediction results, the medium-term prediction results and the long-term prediction results. The short-term prediction result is obtained by predicting the average value of the basic operating parameters in a recent period of time, such as 15 minutes, and the medium-term prediction result is obtained by The prediction is obtained by learning the changing pattern of the basic operating parameters of the long short-term memory network in the past few hours, such as 1 hour, and the long-term prediction result is obtained by time series analysis method according to the changing rules of the daily and weekly cycles in the historical basic operating parameters, such as one day or one week; at the same time, according to the above description, the dynamic threshold interval = [dynamic reference value - (mean of error - lower limit of confidence interval), dynamic reference value + (upper limit of confidence interval - mean of error)], where the confidence interval is determined based on statistical principles, for example, the confidence interval is [mean of error - 3×standard deviation of error, mean of error + 3×standard deviation of error].

[0086] S2. Compare the basic operating parameters of the central air-conditioning terminal obtained in real time with the dynamic benchmark values. When it is detected that the basic operating parameters deviate from the dynamic benchmark values, set an anomaly detection method to fuse the associated features of the scene dimension parameters and the basic operating parameters to form a fusion feature set. Based on the fusion feature set, verify the distribution properties, local performance and scene requirements of the basic operating parameters to identify the anomaly type and locate the source of the anomaly.

[0087] The scene dimension parameters and basic operation parameters belong to different data types and need to be converted into associable feature forms through preprocessing to provide a basis for fusion. The scene dimension parameters are structured and analyzed through the previous steps. The obtained thermal height coefficient, heat flow direction identifier, load adjustment factor and topological influence matrix are used as scene features, and feature extraction is performed on the basic operation parameters. The mean, fluctuation amplitude and change rate of parameters such as temperature and air volume in a short time window are calculated, that is, the rising or falling trend of temperature within a fixed time. The instantaneous value is converted into an operation feature that reflects the change trend. The dimensional difference of different features is eliminated through standardization to ensure that the scene features and operation features can be calculated in the same dimension; thereby, the unstructured scene dimension parameters and the dynamically changing basic operation parameters are converted into standardized features, which solves the problem that the two types of parameters cannot be directly associated due to type differences, and provides a unified data basis for subsequent correlation analysis.

[0088] A single feature cannot reflect the intrinsic connection between the scene and the operating status of the central air-conditioning terminal. It is necessary to explore the correlation between the two to form a fusion feature that can reflect the physical laws; the thermal height coefficient and the heat flow direction identifier are combined with the temperature and air volume characteristics respectively to generate spatial distribution characteristics, which directly associate the spatial attributes and time attributes of the scene with the distribution laws of the basic operating parameters; the load adjustment factor is combined with the water flow and valve opening characteristics to generate scene matching features to associate the scene demand with the actual load regulation capability; at the same time, the topological influence matrix is ​​combined with the water flow characteristics to generate a conductive coupling feature to reflect the degree of mutual influence of water flow between adjacent terminals and reflect the impact of the opening and closing of the bypass pipe on the stability of the terminal operating parameters; the fusion features formed by correlation mining break through the limitations of single parameter analysis, and can directly reflect the physical logic of how the scene dimension parameters affect the basic operating parameters, providing a judgment basis with clear physical meaning for subsequent verification.

[0089] The mined associated features are redundant, that is, some features reflect the same rules. Key features need to be retained through screening and integrated according to the verification dimension to form a structured fusion feature set to ensure that subsequent verification is carried out efficiently; the contribution of each associated feature is reversely verified based on historical abnormal data, the features that have a significant impact on abnormal judgment are retained, and the features with low contribution are eliminated; the screened features are classified according to the verification dimension to form three feature subsets, including a spatial distribution feature subset for distribution property verification, a conduction coupling feature subset for local performance verification, and a scene matching feature subset for scene requirement verification. The three subsets together constitute a fusion feature set, and each feature is associated with a corresponding terminal identifier and timestamp to ensure traceability; redundant features are removed through screening to reduce the computational complexity of subsequent verification, and features are integrated according to the verification dimension to make the fusion feature set correspond one-to-one with the verification logic of distribution properties, local performance and scene requirements, ensuring that subsequent verification steps can directly call relevant features, thereby improving the efficiency and pertinence of anomaly detection.

[0090] In the central air-conditioning system, the distribution law and rationality characteristics of the basic operating parameters of each terminal in the vertical space and time dimensions of the building are specifically reflected in the inherent distribution law of the basic operating parameters in the thermal characteristic area formed by the vertical position of the terminal; the spatiotemporal constraints that the basic operating parameters should meet in the corresponding thermal characteristic area under the influence of seasonal operation mode and time period; the group distribution consistency of the basic operating parameters of the terminals on the same floor through density clustering, and the degree to which individual parameters deviate from the cluster center or spatiotemporal constraints; in summary, the distribution properties are a comprehensive property that reflects whether the basic operating parameters of the central air-conditioning terminals conform to the thermal laws of the building and scenario requirements in terms of vertical space distribution, temporal dynamic changes and group consistency.

[0091] Furthermore, if Figure 3As shown, the verification logic of the distribution property includes:

[0092] The vertical position of the central air conditioning terminal is obtained in real time. Based on the building structure and the location area of ​​the central air conditioning terminal, different floors are divided into thermal characteristic areas. The thermal characteristic areas include low temperature concentration area, heat buffer area, standard area and heat concentration area.

[0093] Extract the heat flow direction identifier encoded according to the seasonal operation mode, and determine the spatiotemporal constraints of the basic operation parameter distribution based on the thermal characteristic area;

[0094] The basic operating parameters of central air-conditioning terminals on the same floor are density clustered, outlier terminals that deviate from the cluster center are automatically identified, and the residuals between the basic operating parameters of the outlier terminals and the spatiotemporal constraints are calculated. When the residuals are not within the dynamic threshold range, it is determined that there is an abnormal distribution property.

[0095] The thermal distribution of central air conditioning on different floors of a building is affected by factors such as vertical height, sunlight and human activities. Dividing the thermal characteristic areas can clarify the thermal characteristics of each floor and provide a basis for judging whether the parameter distribution is reasonable; extracting the spatial distribution feature subset, obtaining the terminal vertical position of the central air conditioning terminal in real time, combining the floor information of the building structure, determining the floor where the central air conditioning terminal is located, and based on the obtained terminal position area, analyzing based on the principle of building thermodynamics, the underground area is divided into a low-temperature gathering area because cold air is easy to accumulate; in the above-ground area, the bottom floor is close to the ground and people enter. The high-rise area is prone to heat accumulation due to factors such as rising hot air and sunlight, and is divided into a heat accumulation area. During the division process, historical temperature data and architectural design information are referred to to ensure that the regional division conforms to the actual thermal distribution law; through the scientific division of thermal characteristic areas, the detection device can understand the thermal characteristics of different floors in more detail, provide an accurate basis for the subsequent determination of the temporal and spatial constraints of the basic operating parameter distribution, and help improve the accuracy of the judgment of parameter distribution anomalies.

[0096] Thermal characteristic zones in different seasons and floors have different requirements for the distribution of basic operating parameters. Determining spatiotemporal constraints can provide a standard for determining whether the actual parameter distribution is normal. Heat flow direction identifiers encoded according to seasonal operating modes are extracted and analyzed in conjunction with the divided thermal characteristic zones. In cooling mode, heat accumulation zones require relatively low temperatures and reasonable temperature gradients. A reasonable temperature range and an allowable fluctuation range over time are set for these zones based on historical data and thermodynamic principles. Low-temperature accumulation zones require temperatures not to be too low to avoid energy waste and equipment loss, and corresponding temperature ranges and fluctuation ranges are also set. In heating mode, the temperature requirements for each zone are opposite. Furthermore, considering the varying human activity and equipment usage at different times of the day, the detection device further refines the constraints based on the time dimension. For example, when office areas are densely populated during the daytime, the temperature and air volume requirements differ from those at night, thus forming a complete set of spatiotemporal constraints for the distribution of basic operating parameters. Clear spatiotemporal constraints provide clear criteria for determining whether the distribution of basic operating parameters is abnormal, enabling the detection device to make judgments based on scientific and reasonable evidence, reducing the possibility of misjudgments and improving the accuracy of anomaly detection.

[0097] By identifying outlier terminals and calculating their residuals relative to the spatiotemporal constraints, it is possible to determine whether there are anomalies in the floor distribution of basic operating parameters and promptly detect potential anomalies. The basic operating parameters of central air conditioning terminals on the same floor are density-clustered. During the clustering process, central air conditioning terminals with similar basic operating parameters are grouped into the same cluster using the concept of density reachability. Outlier terminals that deviate from the cluster center are automatically identified, and the residuals of the basic operating parameters of the outlier terminals relative to the spatiotemporal constraints are calculated. This involves comparing the actual parameter values ​​of the outlier terminals with the reasonable parameter range specified in the spatiotemporal constraints. When the residual is not within the dynamic threshold range, it is determined that a distribution anomaly exists. This allows for accurate identification of terminal devices with abnormal parameters in floor distribution, prompting the detection of potential anomalies. This provides important clues for subsequent anomaly type identification and anomaly source location, helping to improve the stability and reliability of the detection device. The determined distribution anomaly results are an important basis for subsequent comprehensive analysis of anomaly types and location of anomaly sources. If a distribution anomaly exists, the detection device will further combine the results of local performance verification and scenario requirement verification to comprehensively determine the anomaly type and locate the anomaly source.

[0098] In a central air-conditioning system, the operating characteristics and mutual influence of a single or a group of adjacent terminal devices in a local hydraulic system are specifically reflected in the hydraulic coupling relationship formed by the terminals connected by bypass pipes; the conduction law of basic operating parameters between adjacent terminals; the degree of deviation between the theoretical parameter values ​​and the actual operating parameters of adjacent terminals on the conduction path, and whether the deviation meets the constraints of the dynamic threshold range; in summary, local performance is a comprehensive property that reflects the rationality of parameter transmission of central air-conditioning terminal devices in the local hydraulic system, the correlation of mutual influence between terminals, and the consistency between actual operating status and theoretical expectations.

[0099] Furthermore, if Figure 4 As shown, the local performance verification sub-logic includes:

[0100] The topological influence matrix constructed based on the bypass pipe status and the relationship between adjacent terminals is converted into a hydraulic topology graph. The central air conditioning terminals are used as nodes. The connection relationship between the central air conditioning terminals represents the edge of the hydraulic topology graph. The edge weight represents the hydraulic coupling strength in the topological influence matrix. The valve opening of the bypass pipe is obtained in real time to dynamically update the edge weight of the hydraulic topology graph.

[0101] When it is detected that the basic operating parameters of the central air-conditioning terminal deviate from the dynamic reference value, the conduction path between the central air-conditioning terminal and the adjacent terminal is determined by the edges and edge weights of the hydraulic topology graph;

[0102] The deviation value is obtained by subtracting the theoretical parameter value of the adjacent ends in the conduction path from the actual basic operating parameter. When the deviation value is not within the dynamic threshold range, it is determined that there is a local performance abnormality.

[0103] The hydraulic connection relationship between central air-conditioning terminals and the state of the bypass pipe will affect the performance of each central air-conditioning terminal. Constructing a hydraulic topology diagram can intuitively present this relationship, and real-time updates can adapt to changes in the operating state of the central air-conditioning, providing a basis for analyzing the local performance of the terminal; extracting the conductive coupling feature subset, converting the topological influence matrix constructed based on the bypass pipe state and the relationship between adjacent terminals into a visual hydraulic topology diagram, with the central air-conditioning terminal as the node, and the connection relationship between the terminals representing the edge of the hydraulic topology diagram. The weight of the edge is determined according to the hydraulic coupling strength in the topological influence matrix. If the element value corresponding to two terminals in the topological influence matrix is ​​1, it means that they have a hydraulic coupling relationship. In the hydraulic topology diagram, there will be an edge connecting the two terminal nodes, and the weight of the edge is According to the setting of the coupling strength, the valve opening of the bypass pipe is obtained in real time. When the valve opening changes, the hydraulic coupling strength between the relevant terminals will be recalculated, and the weight of the edge of the hydraulic topology map will be dynamically updated. In addition, the detection device will regularly check the connection status of the central air-conditioning terminal. If a new connection or disconnection is found, the structure of the hydraulic topology map will be updated in time to ensure that it accurately reflects the hydraulic connection status of the current system; the hydraulic topology map intuitively displays the hydraulic connection relationship and the degree of mutual influence between the central air-conditioning terminals, so that the hydraulic topology map can reflect the changes in the operating status of the central air-conditioning in real time. This provides a clear and intuitive tool for the detection device to analyze the local performance of the terminal and judge the abnormal propagation path, which helps to improve the efficiency and accuracy of abnormal diagnosis.

[0104] When a basic operating parameter of a central air conditioning terminal is detected to deviate from a dynamic baseline value, the transmission path between the terminal and the adjacent terminals is determined. This helps analyze whether the parameter deviation is caused by the influence of adjacent terminals or hydraulic imbalance of the central air conditioning, thereby determining whether local performance is abnormal. When a basic operating parameter of a central air conditioning terminal is detected to deviate from the dynamic baseline value, the transmission path between the central air conditioning terminal and the adjacent terminals is determined by edges and edge weights in the hydraulic topology graph. Starting from the terminal node that deviates from the basic operating parameter, the transmission path between the terminals is searched based on the connection relationship and weight of the edges. Edges with larger weights are preferentially selected as the main transmission path because the hydraulic influence between adjacent terminals on these transmission paths is more significant. In the process of determining the transmission path, a graph search algorithm is used to gradually explore along the transmission path until all adjacent terminal nodes that affect the basic operating parameter of the central air conditioning terminal are found, forming a complete transmission path. At the same time, the changes in the basic operating parameters of each node along the transmission path are recorded for subsequent analysis. Accurately determining the transmission path helps the detection device quickly analyze the cause of the parameter deviation and determine whether it is the terminal itself that is abnormal, the influence of adjacent terminals, or the hydraulic imbalance of the central air conditioning. This provides key clues for further determining whether local performance is abnormal, helping to improve the accuracy and specificity of abnormality diagnosis.

[0105] By calculating the deviation value between the actual basic operating parameters of the conduction path and the adjacent terminals and comparing it with the dynamic threshold interval, it is possible to determine whether the performance of the terminal in the local hydraulic system is normal and to promptly discover local performance anomalies. The theoretical parameter values ​​of each adjacent terminal on the determined conduction path are subtracted from the actual basic operating parameters to obtain the deviation value. That is, for the temperature parameter of a certain adjacent terminal on the conduction path, the difference between its theoretical temperature value and the actual measured temperature value is calculated as the deviation value of the parameter. The calculated deviation value is then compared with the dynamic threshold interval. When the deviation value is not within the dynamic threshold interval, it is determined that there is a local performance anomaly. In the process of calculating the deviation value and making the judgment, the deviation of multiple parameters is comprehensively considered. As well as the mutual influence relationship between each central air-conditioning terminal on the conduction path, avoiding misjudgment due to a single parameter or simple comparison; thereby being able to accurately judge whether the performance of the central air-conditioning terminal in the local hydraulic system is normal, and timely discover local performance abnormalities, providing an important basis for subsequent abnormality type identification and abnormal source positioning, helping to improve the reliability and stability of the detection device, and ensure the normal operation of the central air-conditioning system. The determined local performance abnormality result is one of the important bases for subsequent comprehensive analysis of the abnormality type and positioning of the abnormal source. If there is a local performance abnormality, the detection device will combine the results of distribution property verification and scenario demand verification to comprehensively judge the abnormality type and locate the abnormal source so as to take corresponding treatment measures.

[0106] Furthermore, the verification sub-logic of scenario requirements includes:

[0107] Obtain the occasion type identification of the central air-conditioning terminal in real time, parse the occasion type identification and extract the basic operating parameters under the occasion type identification according to the scene demand database to form a scene demand set;

[0108] The basic operating parameters obtained in real time are matched with the scenario requirement set. If the basic operating parameters do not match the scenario requirement set, it is determined that there is a scenario requirement anomaly.

[0109] Continuously record the verification results and basic operating parameters of each scenario requirement. When there are multiple abnormal scenario requirements, analyze the historical basic operating parameters to optimize the scenario requirement database.

[0110] Different usage scenarios have specific requirements for the operating parameters of central air conditioners. Only by generating the corresponding scenario requirement set can we provide an accurate reference standard for subsequent judgment on whether the current operating parameters meet the scenario requirements; obtain the scenario type identification of the central air conditioner terminal in real time, and parse it through the semantic parsing algorithm after obtaining the scenario type identification. According to the parsed results, accurate retrieval is performed in the scenario requirement database, which stores a large number of basic operating parameter standards for different occasions under various circumstances. These standards are established by collecting industry specifications and a large amount of historical operating data summary. The retrieved relevant parameters are extracted to form a scenario requirement set for the current occasion type; by generating a scenario requirement set, the central air conditioner terminal detection device can clearly understand the specific operating requirements in different scenarios, laying a solid foundation for achieving precise control and efficient operation, which helps to improve user comfort in different scenarios, and also avoids energy waste caused by unreasonable parameter settings.

[0111] By matching and judging the basic operating parameters obtained in real time with the scene requirement set, it is possible to promptly discover whether the current operating parameters meet the actual requirements of the scene, thereby identifying potential anomalies caused by scene adaptation problems; extracting the scene matching feature subset, matching and judging the basic operating parameters of the central air-conditioning terminal obtained in real time with the various parameter standards in the generated scene requirement set one by one. When the basic operating parameters do not match the scene requirement set, it is determined that there is a scene requirement anomaly, and record relevant information, including the name, actual value, standard value of the abnormal parameter, and the time when the anomaly occurred; through precise parameter matching judgment, the difference between the central air-conditioning operating parameters and the scene requirements can be quickly and accurately discovered, and anomalies caused by scene adaptation problems can be warned in time, ensuring that the detection device can operate stably and efficiently in various scenarios, thereby improving the reliability of the detection device and user satisfaction.

[0112] Over time and as actual usage changes, the demand for central air conditioning in different scenarios will also change. Continuously recording the results of scenario demand verification and basic operating parameters, and optimizing the database, can make the scenario demand database more aligned with actual needs and improve the adaptability and accuracy of the detection device. The results of each scenario demand verification and the corresponding basic operating parameters are continuously recorded. When multiple scenario demand anomalies are detected, the optimization process is initiated. First, an in-depth analysis of the historical basic operating parameters is conducted. Using data mining and machine learning algorithms, the patterns and potential influencing factors of the anomalies are identified. In the example of a shopping mall, analysis revealed that the sudden increase in customer traffic and changes in equipment load during promotions were the main causes of temperature and air volume anomalies. Based on the analysis results, the parameter standards for the mall promotion scenario in the scenario demand database were adjusted, appropriately reducing the temperature setting range and increasing the sensitivity of air volume adjustment. During the optimization process, reference is made to successful experiences and data from other similar scenarios to ensure the scientific and reasonableness of the optimized parameter standards. After the optimization is completed, the new parameter standards are updated to the scenario demand database, and the updated results are continuously monitored, with further adjustments and improvements based on actual feedback.

[0113] By optimizing the scenario demand database, central air conditioners can better adapt to the ever-changing needs in different scenarios and improve the intelligence and adaptability of the detection device. This not only helps to improve user experience, but also further reduces energy consumption and achieves the goal of energy conservation and emission reduction. The optimized scenario demand database provides more accurate and more practical parameter standards for subsequent scenario demand verification, which can improve the accuracy of parameter matching judgment, reduce the occurrence of misjudgment and missed judgment, thereby more effectively ensuring the stable operation of the detection device and providing a more reliable basis for abnormal diagnosis and processing.

[0114] Specifically, the logic for locating the anomaly source includes:

[0115] Calculate the outliers for distribution anomalies, local performance anomalies, and scenario demand anomalies respectively, and dynamically assign verification weights for distribution anomalies, local performance anomalies, and scenario demand anomalies based on the scenario type to fuse them to obtain the total value of dimensional anomalies.

[0116] Based on the hydraulic topology, the node orientation is determined based on the total value of dimensional anomalies to form the direction of anomaly propagation. The hydraulic coupling strength is also used to determine the probability of anomaly propagation on each edge. Furthermore, the central air conditioning terminals with anomalies are density clustered based on the total value of dimensional anomalies to identify anomaly clusters.

[0117] If the abnormal cluster includes a single central air-conditioning terminal, the abnormal source is determined to be a terminal monomer and the attribute characteristic of the abnormal source is local failure; if the abnormal cluster includes multiple central air-conditioning terminals, and it is judged that the edges with high probability of abnormal propagation are densely distributed in the abnormal cluster, the abnormal source is determined to be a group and the attribute characteristic of the abnormal source is cascade risk.

[0118] Distribution anomalies, local performance anomalies and scenario demand anomalies reflect the problems existing in the central air-conditioning terminals from different angles. By calculating these anomalies separately and performing weighted fusion, a comprehensive dimensional anomaly total value can be obtained, thereby more comprehensively and accurately evaluating the degree of abnormality of the detection device and providing a more reliable basis for locating the abnormal source; first, the anomaly values ​​of distribution anomalies, local performance anomalies and scenario demand anomalies are calculated separately. For distribution anomalies, the anomaly value is determined based on the residuals between the basic operating parameters of the outlier terminal and the temporal and spatial constraints. The larger the residual, the higher the degree of distribution anomaly and the corresponding larger the anomaly value. For local performance anomalies, the anomaly value is determined based on the deviation value between the actual basic operating parameters of the conduction path and the adjacent terminal. The more the deviation value exceeds the dynamic threshold interval, the higher the local performance anomaly value. For scenario demand anomalies, the anomaly value is determined based on the number of mismatched parameter items and the degree to which the parameters deviate from the standard. The more mismatched items and the greater the deviation, the larger the scenario demand anomaly value.

[0119] Then, the verification weights of these three anomalies are dynamically assigned according to the current occasion type. In places with extremely high requirements for environmental stability, such as hospitals, the weights of distribution property anomalies and scenario demand anomalies will be relatively high. In some industrial plants and other places that pay more attention to the stability of the hydraulic system, the weight of local performance anomalies will be increased. The three calculated anomaly values ​​are weighted and summed according to the assigned weights to obtain the total value of dimensional anomalies; through anomaly value calculation and weight fusion, the anomaly information of multiple dimensions is integrated to form a comprehensive evaluation index, avoiding the one-sidedness and misjudgment caused by judging only based on a single dimensional anomaly, and can more comprehensively and accurately reflect the overall abnormal condition of the central air-conditioning, providing more scientific and reliable basic data for the subsequent positioning of the anomaly source.

[0120] Based on the hydraulic topology map, anomaly propagation analysis and anomaly cluster identification are carried out to clearly present the propagation path and range of anomalies in central air conditioning, find out the areas where parameter anomalies are closely related, thereby narrowing the scope of investigation of anomaly sources and improving the efficiency and accuracy of locating anomaly sources; based on the hydraulic topology map, the propagation direction of the anomaly is determined according to the direction of the node with higher dimensional anomaly total value. If the dimensional anomaly total value of terminal A is significantly higher than that of other terminals, and is connected to terminals B and C through edges in the hydraulic topology map, it will be preliminarily judged that the anomaly propagates from terminal A to terminals B and C. At the same time, combined with the weight of the edges in the hydraulic topology map, the anomaly propagation probability of each edge is calculated. The larger the weight of the edge, the closer the hydraulic connection between the two terminals, and the higher the probability of the anomaly propagating through the edge; then based on the dimensional anomaly total value Density clustering is performed on central air-conditioning terminals with abnormalities. During the clustering process, terminals with similar total dimensional anomaly values ​​and adjacent positions in the hydraulic topology map or with strong hydraulic connections are divided into the same anomaly cluster through a clustering algorithm. In a certain area, if the total dimensional anomaly values ​​of multiple terminals are greater than the set anomaly threshold and they are closely connected to each other in the hydraulic topology map, these terminals are identified as an anomaly cluster; through anomaly propagation analysis and anomaly cluster identification, the propagation path and impact range of the anomaly in the central air-conditioning can be intuitively displayed, the area where parameter anomalies are concentrated can be quickly locked, and the scope of the anomaly source investigation can be narrowed from the entire central air-conditioning to a specific anomaly cluster, which greatly improves the efficiency of anomaly source positioning, reduces the time and cost of anomaly investigation, and also provides a clear direction for subsequent anomaly handling.

[0121] Accurately determining the attribute characteristics of the abnormal source can provide a basis for the subsequent targeted scenario response strategy, ensure that the abnormality is handled quickly and effectively, and reduce the impact on the normal operation of the central air-conditioning; determine the type and attribute characteristics of the abnormal source according to the characteristics of the abnormal cluster. If the abnormal cluster only contains a single central air-conditioning terminal, and after analysis it is found that the abnormality of the terminal has no obvious impact on the adjacent terminals, that is, there is no sign of abnormal propagation, then the abnormal source is determined to be a terminal monomer, and the attribute characteristics of the abnormal source are local failure. Preferably, the valve opening of a certain terminal is abnormal, but the operating parameters of the surrounding terminals are not disturbed. This situation meets the characteristics of local failure of the terminal monomer.

[0122] If an anomaly cluster contains multiple central air-conditioning terminals, and analysis shows that edges with high anomaly propagation probability are densely distributed within the anomaly cluster, indicating a strong mutual influence relationship between these terminals and a cascade propagation trend of the anomaly within the cluster, the anomaly source will be determined to be a group, and the attribute characteristic of the anomaly source is cascade risk. Preferably, temperature anomalies occur simultaneously at multiple terminals on a floor, and the hydraulic coupling strength between them is high. Through anomaly propagation analysis, it is found that the anomaly gradually spreads from one terminal to other terminals. This situation belongs to group cascade risk anomaly.

[0123] Accurate determination of the attribute characteristics of the abnormal source can enable the detection device to formulate the most reasonable and effective scenario response strategy for abnormalities of different types and attribute characteristics, avoiding the waste of resources and poor processing results caused by blindly handling abnormalities. This helps to improve the efficiency and quality of abnormality handling, restore the normal operation of the central air-conditioning as soon as possible, and ensure the user's usage needs; the determined abnormality type and abnormal source attribute characteristics are the direct basis for matching the scenario response strategy, and at the same time provide a clear direction and goal for subsequent monitoring of basic operating parameter changes and updating the compensation rule library.

[0124] S3. Extract the attribute characteristics of the abnormal type and abnormal source, match the scenario response strategy and control the central air-conditioning terminal detection device to adjust the central air-conditioning terminal, and at the same time monitor the change trend of each basic operating parameter in the central air-conditioning terminal to update the compensation rule base.

[0125] Specifically, the matching logic of the scenario response strategy includes:

[0126] Establish scenario response strategies, classify them into three levels according to the anomaly type, anomaly source, and attribute characteristics of the anomaly source, and set an index identifier for each scenario response strategy;

[0127] Obtain the identification results of the attribute characteristics of the anomaly type and anomaly source in real time, and match them in the scenario response strategy according to the three-level classification. If multiple scenario response strategies are matched, the priority is determined according to the scenario type identifier;

[0128] Execute the matching scenario response strategy and monitor the changing trend of each basic operating parameter in the central air-conditioning terminal in real time to provide feedback and modify the scenario response strategy.

[0129] Central air conditioning anomalies are complex and diverse, with different handling methods corresponding to different anomaly types, sources, and attribute characteristics. A three-level classification and indexing approach enables systematic management of scenario response strategies, facilitating rapid retrieval and matching, and improving anomaly handling efficiency. Common central air conditioning anomalies are comprehensively sorted out and categorized by anomaly type, source, and attribute characteristics. Anomaly types include equipment anomalies, system imbalances, and false alarms. Anomaly sources include single terminals and groups. Attribute characteristics of anomaly sources include local failure and cascading risk. A corresponding scenario response strategy is developed for each classification combination. For a local failure of a single terminal, the scenario response strategy is to shut down the anomalous terminal and activate a backup terminal. For a group cascade risk, the scenario response strategy is to adjust the group's shared valve and coordinate load adjustments in adjacent groups. A unique index identifier is assigned to each scenario response strategy. This identifier, in coded form, contains characteristic information about the anomaly type, source, and attribute characteristics, facilitating rapid identification and recall. This creates a structured and standardized scenario response strategy library, making anomaly handling strategies clear and explicit, facilitating rapid location and execution of appropriate scenario response strategies, reducing manual intervention and decision-making time, and improving the timeliness and accuracy of anomaly responses.

[0130] After an exception occurs, it is necessary to quickly match an accurate processing strategy from the scenario response strategy library. When there are multiple scenario response strategies matched, the priority is determined according to the scenario type, which can ensure that the strategy most suitable for the current scenario is executed first, thereby improving the exception handling effect; the attribute feature results of the identified exception type and the exception source are obtained in real time, and accurate matching is performed in the scenario response strategy library according to the three-level classification system. When there are multiple scenario response strategies matched, the priority is determined according to the scenario type identification obtained in real time. Preferably, in the hospital scenario, due to the extremely high requirements for environmental stability, the strategy that can restore the stability of the central air conditioning the fastest is given priority, while in the office area during the low-load period at night, the strategy with lower energy consumption is given priority. Taking into account factors such as the personnel density, usage time and special needs of the scenario, the matched scenario response strategies are sorted and the strategy with the highest priority is selected as the execution plan; thereby ensuring that in complex exception situations, the response strategy most suitable for the current scenario can be selected quickly and accurately, avoiding invalid processing or negative effects caused by blindly executing strategies, improving the pertinence and effectiveness of exception processing, and ensuring the stable operation of central air conditioning in different scenarios.

[0131] Executing the matching scenario response strategy to handle anomalies and providing feedback and correction strategies through real-time monitoring of parameter changes can ensure that the strategy effectively resolves anomalies and continuously optimizes the scenario response strategy to adapt to different abnormal situations and operating scenarios. The determined scenario response strategy is converted into specific control instructions and sent to the central air-conditioning terminal detection device for execution. If the strategy is to adjust the valve opening of a certain group, an opening adjustment instruction is sent to the corresponding valve. During the execution of the scenario response strategy, the changes in each basic operating parameter in the central air-conditioning terminal are continuously monitored in real time. The acquired data is compared and analyzed with the dynamic benchmark value and scenario requirements to judge the execution effect of the scenario response strategy. If the basic operating parameters fail to return to the normal range or new abnormal situations occur after the scenario response strategy is executed, the scenario response strategy will be fed back and corrected according to the deviation. The correction method includes adjusting the parameter settings in the scenario response strategy and changing the execution step order, or even reselecting other matched scenario response strategies. After each correction, the correction content and the parameter changes before and after the correction are recorded for subsequent strategy optimization.

[0132] This enables dynamic adjustment and optimization of scenario response strategies, ensuring that scenario response strategies can effectively resolve anomalies, improving the adaptability and anomaly handling capabilities of detection devices, and continuously improving the scenario response strategy library through continuous feedback correction to better cope with various complex abnormal scenarios and ensure stable and efficient operation of central air conditioning. The change data of basic operating parameters monitored during the execution of scenario response strategies is an important basis for subsequent updating of the compensation rule library. The changes in basic operating parameters reflect the difference between the current operating state of the central air conditioning and the ideal state, providing actual data support for analyzing the correlation between scenario dimension parameters and deviations from basic operating parameters.

[0133] Specifically, the update logic of the compensation rule base includes:

[0134] Real-time monitoring of the deviation of the basic operating parameters of the central air-conditioning terminal from the dynamic reference value, including the deviation amplitude and degree, and judging whether to update the compensation rule base according to the deviation;

[0135] Analyze the correlation between scenario dimension parameters and deviations from basic operating parameters to generate candidate rules, simulate the verification effect of executing candidate rules when basic operating parameters deviate from dynamic baseline values, and select compensation rules based on the verification effect;

[0136] The filtered compensation rules are added to the compensation rule library, and priorities are assigned according to the applicable frequency of the compensation rules. The changing trend of each basic operating parameter in the central air-conditioning terminal is monitored in real time to iteratively update the compensation rule library.

[0137] The operating environment of central air conditioners is complex and changeable. The deviation of basic operating parameters from dynamic benchmark values ​​means that the existing compensation rules are no longer applicable. Real-time monitoring of deviations and judgment on whether to update the compensation rule library can timely discover potential problems in the operation of central air conditioners and ensure the effectiveness of compensation rules. The basic operating parameters of the central air conditioner terminal are obtained in real time and continuously compared with the dynamic benchmark values ​​to calculate the deviation amplitude and degree of deviation of the basic operating parameters. The deviation amplitude reflects the difference between the actual value of the parameter and the dynamic benchmark value. The degree of deviation is combined with historical data and scenario requirements to evaluate the severity of the deviation. If the actual temperature value is higher than the dynamic benchmark value, and the deviation is in the current scenario, If the dynamic threshold range has been exceeded, the deviation is recorded. When the monitored deviation meets the preset update trigger conditions, that is, the same parameter deviates multiple times within a certain period of time and the deviation is large or serious deviation occurs, it is determined that the compensation rule library needs to be updated and the update process is started. These trigger conditions are pre-set based on a large amount of historical operating data and industry experience, and can be adjusted according to actual operating conditions; thereby, abnormal deviations of basic operating parameters can be discovered in a timely manner, and it can be accurately determined whether the compensation rule library needs to be updated, avoiding the decline in central air-conditioning operating efficiency or abnormality due to lagging compensation rules, and improving the stability and reliability of the detection device.

[0138] Deviations in basic operating parameters are often correlated with scenario-dimensional parameters. Analyzing the correlation between the two and generating candidate rules can identify the root cause of parameter deviations, providing a scientific basis for optimizing compensation rules and making them more aligned with actual operating needs. The scenario-dimensional parameters and deviation data for the basic operating parameters at the time of the current parameter deviation are obtained and integrated. Using an association rule mining algorithm, potential correlations between the scenario-dimensional parameters and deviations in basic operating parameters are identified. For example, if it is found that the water flow at the central air conditioning terminal is prone to abnormal deviations when the bypass valve opening reaches a certain level and the adjacent terminal is under high load, this correlation is used as the basis for candidate rules. Based on the analyzed correlations, multiple candidate rules are generated. Each candidate rule describes how basic operating parameters should be adjusted to reduce deviations under specific scenario-dimensional parameters. These candidate rules cover different scenarios and parameter combinations, providing a rich selection for subsequent screening of appropriate compensation rules. This in-depth exploration of the intrinsic connection between scenario-dimensional parameters and deviations in basic operating parameters allows the generated candidate rules to be targeted and scientific, providing strong support for updating the compensation rule library, helping to improve the accuracy and effectiveness of compensation rules and enabling the detection device to better adapt to different operating scenarios.

[0139] Filter out effective rules from the candidate rules and add them to the compensation rule library. At the same time, assign priorities and perform iterative updates based on the applicable frequency, which can ensure the practicality and advancement of the compensation rule library and enable the detection device to provide accurate compensation guidance when facing various operating conditions; simulate the generated candidate rules, reproduce the scenario where the basic operating parameters deviate from the dynamic benchmark value in the simulation environment, execute the candidate rules, monitor the changes in the basic operating parameters, and filter out the rules that can effectively reduce the parameter deviation and make the central air-conditioning operation more stable from the candidate rules based on the simulation results. If a candidate rule can reduce the temperature deviation in the simulation without negatively affecting other parameters, then the rule is selected. After the candidate rules are selected, the screened compensation rules are added to the compensation rule library, and priorities are assigned to them according to the applicable frequency of the compensation rules. Compensation rules with high applicable frequency have higher priorities and will be called first when calculating the baseline value offset. At the same time, the changes in each basic operating parameter in the central air-conditioning terminal are continuously monitored in real time, and the actual effect data after the execution of the compensation rules are collected. When it is found that the existing compensation rules are not effective in actual operation, or new scenarios and parameter deviations occur, the association relationship is re-analyzed and candidate rules are generated. The compensation rule library is iteratively updated, invalid rules are deleted, existing rules are optimized, and new rules are added to ensure that the compensation rule library is always kept in the best state.

[0140] Through strict rule screening, reasonable priority allocation and continuous iterative updates, the accuracy and effectiveness of the compensation rule library are guaranteed. The basic operating parameters can be flexibly adjusted according to the actual operating conditions, the operating efficiency and stability of the central air conditioner can be improved, and the probability of abnormalities can be reduced. The updated compensation rule library will be used to generate the baseline value offset. More accurate and effective compensation rules can make the calculation of the baseline value offset more in line with actual operating needs, thereby improving the accuracy of the dynamic baseline value and dynamic threshold interval, and providing a more reliable basis for subsequent abnormality detection, verification and processing.

Claims

1. A control method for a central air-conditioning terminal detection device, characterized in that: include: Identify whether the regional temperature of the central air-conditioning terminal is abnormal. When the regional temperature is abnormal, verify whether the basic operating parameters of the central air-conditioning terminal are deviated. The basic operating parameters and scenario dimension parameters of the central air conditioning terminal are obtained in real time. Based on the scenario dimension parameters, the baseline value offset of the basic operating parameters is generated through the compensation rule library. The baseline value offset is superimposed through the time series prediction method to determine the dynamic baseline value and dynamic threshold range; Compare basic operating parameters with dynamic benchmark values. In response to deviations from the dynamic benchmark values, set an anomaly detection method to fuse the associated features of the scenario dimension parameters and the basic operating parameters to form a fused feature set. Based on the fused feature set, verify the distribution properties, local performance, and scenario requirements of the basic operating parameters to identify the anomaly type and locate the anomaly source. Extract the attribute characteristics of the abnormal type and abnormal source, match the scenario response strategy and control the central air-conditioning terminal detection device to adjust the central air-conditioning terminal, while monitoring the changing trend of basic operating parameters to update the compensation rule base.

2. A control method for a central air conditioning terminal detection device according to claim 1, characterized in that: The determination logic of the dynamic reference value includes: The basic operating parameters of central air conditioning terminals are predicted at multiple scales, including short-term, medium-term and long-term scales, using the time series prediction method. Fusion weights are assigned according to the time length of the scales. The prediction results of different scales are fused based on the fusion weights to obtain the initial benchmark value. The consistency of the prediction results of different scales is verified, and the prediction results of different scales are corrected by combining historical adjustment data and scene dimension parameters. The baseline value offset and the initial baseline value are superimposed to determine the dynamic baseline value, and the confidence interval of the prediction result is determined based on the error of the historical prediction result to generate the dynamic threshold interval.

3. A control method for a central air conditioning terminal detection device according to claim 2, characterized in that: The generation sub-logic of the reference value offset includes: Obtain scene dimension parameters of central air conditioning terminals in real time, pre-process and analyze them to obtain thermal height coefficient, heat flow direction identifier, load adjustment factor and topology influence matrix; Extract the compensation rule base, determine the vertical thermal compensation amount based on the thermal height coefficient and heat flow direction identifier, determine the occasion load compensation amount based on the load adjustment factor and air volume and water flow, and monitor the valve opening of the bypass pipe to determine the terminal impact compensation amount based on the topological impact matrix and the temperature change of the adjacent terminal; The vertical thermal compensation, occasion load compensation and terminal impact compensation are processed by weighted summation to generate a reference value offset and recorded in the compensation rule base.

4. A control method for a central air-conditioning terminal detection device according to claim 3, characterized in that: The parsing sub-logic of the scene dimension parameters includes: Obtain the vertical height of the central air conditioning terminal in real time, determine the location area of ​​the central air conditioning terminal based on the building structure, and assign a thermal height coefficient to the location area based on the principles of building thermodynamics; Obtain the seasonal operation mode and occasion type identification of the central air-conditioning terminal in real time, encode the seasonal operation mode into the heat flow direction identification, and map the occasion type identification into the load adjustment factor; The bypass pipe status and adjacent terminal relationship of the central air-conditioning terminal are obtained in real time, and a topological influence matrix is ​​constructed with the central air-conditioning terminal as the matrix node. The matrix elements of the topological influence matrix represent the hydraulic coupling strength between the central air-conditioning terminals.

5. A control method for a central air-conditioning terminal detection device according to claim 4, characterized in that: The logic for locating the abnormal source includes: Calculate the outliers for distribution anomalies, local performance anomalies, and scenario demand anomalies respectively, and dynamically assign verification weights for distribution anomalies, local performance anomalies, and scenario demand anomalies based on the scenario type to fuse them to obtain the total value of dimensional anomalies. Based on the hydraulic topology, the node orientation is determined based on the total value of dimensional anomalies to form the direction of anomaly propagation. The hydraulic coupling strength is also used to determine the probability of anomaly propagation on each edge. Furthermore, the central air conditioning terminals with anomalies are density clustered based on the total value of dimensional anomalies to identify anomaly clusters. If the abnormal cluster includes a single central air-conditioning terminal, the abnormal source is determined to be a terminal monomer and the attribute characteristic of the abnormal source is local failure; if the abnormal cluster includes multiple central air-conditioning terminals, and it is judged that the edges with high probability of abnormal propagation are densely distributed in the abnormal cluster, the abnormal source is determined to be a group and the attribute characteristic of the abnormal source is cascade risk.

6. A control method for a central air-conditioning terminal detection device according to claim 5, characterized in that: The verification sub-logic of the distribution property includes: Obtain the vertical position of the central air conditioning terminal in real time, and divide different floors into thermal characteristic areas based on the building structure and the location area of ​​the central air conditioning terminal; Extract the heat flow direction identifier encoded according to the seasonal operation mode, and determine the spatiotemporal constraints of the basic operation parameter distribution based on the thermal characteristic area; The basic operating parameters of central air-conditioning terminals on the same floor are density clustered, outlier terminals that deviate from the cluster center are automatically identified, and the residuals between the basic operating parameters of the outlier terminals and the spatiotemporal constraints are calculated. When the residuals are not within the dynamic threshold range, it is determined that there is an abnormal distribution property.

7. A control method for a central air-conditioning terminal detection device according to claim 6, characterized in that: The local performance verification sub-logic includes: The topological influence matrix constructed based on the bypass pipe status and the relationship between adjacent terminals is converted into a hydraulic topology graph. The central air conditioning terminals are used as nodes. The connection relationship between the central air conditioning terminals represents the edge of the hydraulic topology graph. The edge weight represents the hydraulic coupling strength in the topological influence matrix. The valve opening of the bypass pipe is obtained in real time to dynamically update the edge weight of the hydraulic topology graph. When it is detected that the basic operating parameters of the central air-conditioning terminal deviate from the dynamic reference value, the conduction path between the central air-conditioning terminal and the adjacent terminal is determined by the edges and edge weights of the hydraulic topology graph; The deviation value is obtained by subtracting the theoretical parameter value of the adjacent ends in the conduction path from the actual basic operating parameter. When the deviation value is not within the dynamic threshold range, it is determined that there is a local performance abnormality.

8. A control method for a central air-conditioning terminal detection device according to claim 7, characterized in that: The verification sub-logic required by the scenario includes: Obtain the occasion type identification of the central air-conditioning terminal in real time, parse the occasion type identification and extract the basic operating parameters under the occasion type identification according to the scene demand database to form a scene demand set; The basic operating parameters obtained in real time are matched with the scenario requirement set. If the basic operating parameters do not match the scenario requirement set, it is determined that there is a scenario requirement anomaly. Continuously record the verification results and basic operating parameters of each scenario requirement. When there are multiple abnormal scenario requirements, analyze the historical basic operating parameters to optimize the scenario requirement database.

9. A control method for a central air-conditioning terminal detection device according to claim 8, characterized in that: The matching logic of the scenario response strategy includes: Establish scenario response strategies, classify them into three levels according to the anomaly type, anomaly source, and attribute characteristics of the anomaly source, and set an index identifier for each scenario response strategy; Obtain the identification results of the attribute characteristics of the anomaly type and anomaly source in real time, and match them in the scenario response strategy according to the three-level classification. If multiple scenario response strategies are matched, the priority is determined according to the scenario type identifier; Execute the matching scenario response strategy and monitor the changing trend of each basic operating parameter in the central air-conditioning terminal in real time to provide feedback and modify the scenario response strategy.

10. A control method for a central air-conditioning terminal detection device according to claim 9, characterized in that: The update logic of the compensation rule base includes: Real-time monitoring of the deviation of the basic operating parameters of the central air-conditioning terminal from the dynamic reference value, including the deviation amplitude and degree, and judging whether to update the compensation rule base according to the deviation; Analyze the correlation between scenario dimension parameters and deviations from basic operating parameters to generate candidate rules, simulate the verification effect of executing candidate rules when basic operating parameters deviate from dynamic baseline values, and select compensation rules based on the verification effect; The filtered compensation rules are added to the compensation rule library, and priorities are assigned according to the applicable frequency of the compensation rules. The changing trend of each basic operating parameter in the central air-conditioning terminal is monitored in real time to iteratively update the compensation rule library.

Images