Central air conditioning energy consumption on-line monitoring and diagnosing method

Abstract

The present application relates to air conditioning technology field, and disclose a kind of central air conditioning energy consumption on-line monitoring and diagnosis method, for solving the low precision of the energy waste quantification evaluation of the specific components of central air conditioning in traditional method;The present application, first, sensor is arranged at key node, real-time acquisition multi-dimensional operating parameter and transmission to central processing unit, parameter is classified and compared with pre-set benchmark to construct deviation vector, based on deviation information to identify compressor, condenser, evaporator, fan and water pump and establish component level energy consumption correlation data, on this basis, total energy consumption is split into independent energy consumption component according to start, stable operation and standby phase, and compared with standard energy consumption model point by point, the energy consumption deviation value of each component in each phase is calculated and written into diagnosis matrix, combined with diagnosis result, operating parameter is adjusted in closed loop, so as to realize the on-line quantification evaluation and energy-saving operation optimization of the energy waste of central air conditioning component.

Description

Technical Field

[0001] This invention relates to the field of air conditioning technology, specifically to a method for online monitoring and diagnosis of energy consumption in central air conditioning systems. Background Technology

[0002] With the advancement of building energy conservation and green development, the online monitoring and diagnosis of central air conditioning systems, as a major component of building energy consumption, has become a key technology for improving energy efficiency. Traditional methods for monitoring and diagnosing central air conditioning energy consumption mainly rely on the acquisition of overall system data and empirical formulas for evaluation. For example, they collect total energy consumption parameters (such as electricity, temperature, and flow rate) through sensors and calculate system efficiency based on simplified models. However, these methods have low accuracy in quantifying energy waste in specific components (such as compressors, fans, or heat exchangers), resulting in large deviations in diagnostic results and an inability to accurately identify component-level energy losses, thus affecting energy-saving optimization and fault prevention.

[0003] For example, CN104075403B discloses an air conditioning energy consumption monitoring and diagnosis system and method, including a data acquisition module, a database, and a user terminal. It performs overall energy consumption diagnosis and provides energy-saving suggestions by collecting real-time air conditioning operation data. While this method achieves remote monitoring, it relies on empirical formulas to assess the total system energy consumption. When quantifying energy waste in specific components, it fails to consider internal thermodynamic dynamic changes and interference factors, resulting in insufficient assessment accuracy. It cannot accurately decompose the waste ratio of components such as compressors, and the error can reach over 15% in practical applications, limiting refined diagnosis. Similarly, CN101975673A discloses a real-time energy efficiency monitoring system and method for central air conditioning systems, describing a system including sensors and an energy efficiency coefficient calculation module. It achieves real-time diagnosis by monitoring the total air conditioning load and energy consumption parameters. This method emphasizes energy efficiency coefficient calculation, but it also relies on empirical models for overall assessment. It has limitations in quantifying energy waste in specific components, such as ignoring nonlinear losses during compressor operation and the influence of environmental variables, resulting in coarse quantification results that cannot provide precise waste data at the component level, thus affecting the implementation of targeted optimization measures.

[0004] The low accuracy of quantitative assessment of energy waste in specific components of central air conditioning systems using traditional methods stems from reliance on simplified empirical formulas and overall data analysis, lacking detailed modeling of component-level dynamic thermal balance and multivariate coupling. This not only leads to low diagnostic accuracy but also results in untimely identification of energy waste, increasing system operating costs and maintenance difficulty. Therefore, a novel approach is needed to address this technical problem, improve the accuracy of quantitative assessment, and achieve more effective energy consumption diagnosis and energy-saving management. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for online monitoring and diagnosis of energy consumption in central air conditioning systems, which solves the problem of low accuracy in quantitative assessment of energy waste in specific components of central air conditioning systems in traditional methods.

[0006] To achieve the goals of more effective energy consumption diagnosis and energy-saving management mentioned in the background section, the present invention provides the following technical solution:

[0007] A method for online monitoring and diagnosis of central air conditioning energy consumption includes:

[0008] S1: Collect various operating parameters of the central air conditioning system in real time and send the collected operating parameters to the central processing unit through the data transmission channel;

[0009] S2: The central processing unit performs a classification operation on the transmitted operating parameters and compares the classified operating parameters with the preset benchmark parameters one by one.

[0010] S3: Identify specific components in the central air conditioning system based on the deviation information obtained from the comparison, and associate and match the identified specific components with the corresponding energy consumption data;

[0011] S4: Isolate the independent energy consumption components of a specific component from the associated and matched energy consumption data, and compare and analyze the isolated independent energy consumption components with the predefined standard energy consumption model;

[0012] S5: Calculate the energy consumption deviation of a specific component using the independent energy consumption components after comparative analysis, and integrate the calculated deviation into the overall system diagnostic framework;

[0013] S6: The final output is an integrated diagnostic framework for continuous monitoring and adjustment of the central air conditioning system.

[0014] In a preferred embodiment, various operating parameters of the central air conditioning system are collected in real time, and the collected operating parameters are sent to the central processing unit through a data transmission channel, including:

[0015] Temperature, pressure, flow, power and speed sensors are placed in key locations, using adhesive, threaded, flanged or insert, clamp-on current transformers and magneto-optical mounting.

[0016] The clock of each sensor is synchronized, time synchronization messages are broadcast periodically, and configuration parameters with a fixed sampling frequency are sent out.

[0017] Data is read through polling or priority scheduling, and parameters are recorded in the order of system topology.

[0018] The front-end node packages the data and sends it to the central processing unit via wireless or wired channels, employing encryption algorithms and electromagnetic interference protection measures.

[0019] The receiving side calculates the checksum and processes messages that fail the checksum verification; at the same time, it temporarily stores the data in a buffer queue and uploads it in batches after the network is restored.

[0020] In a preferred embodiment, the central processing unit performs a classification operation on the transmitted operating parameters, comparing each classified operating parameter with a preset benchmark parameter, including:

[0021] Records are read from the receive buffer in chronological order, and parameter identifiers, values, and timestamps are extracted. The parameters are then classified according to their physical attributes and their role in the system.

[0022] Bind category labels to each parameter and update the mapping table to complete the classification when a new monitoring quantity is added;

[0023] The categorized operating parameters are compared one by one with the preset benchmark parameters;

[0024] Construct a unified measure of relative deviation and compare the actual values ​​of each parameter with their respective benchmark values;

[0025] The various deviations are summarized to form a deviation vector organized by category.

[0026] In a preferred embodiment, identifying specific components in the central air conditioning system based on the deviation information obtained from the comparison includes:

[0027] Read the normalized deviation values, timestamps, parameter identifiers, and category information of thermodynamic, mechanical, and control deviation vectors from the deviation basis;

[0028] Aggregate deviation vectors according to parameter category and system topology location, filter deviation points whose normalized deviation exceeds a preset threshold, and combine them into deviation pattern units.

[0029] The system calls a preset deviation mapping table to identify components and stores component identifiers, location regions, and typical deviation pattern combinations.

[0030] Traverse the deviation patterns within the current time window, match the features in the deviation mapping table, and select the component identifier based on pattern weight, deviation magnitude, and duration.

[0031] In a preferred embodiment, the identified specific components are associated and matched with corresponding energy consumption data, including:

[0032] After the component is identified, it will be associated and matched with the corresponding energy consumption data.

[0033] The energy consumption time is aligned using the deviation timestamp as an index, and the energy consumption data is corrected by interpolation or expanding the window when the alignment threshold is exceeded.

[0034] Build an associated dataset containing deviation information and energy consumption curves for each component;

[0035] The total power is allocated according to the operating status and rated proportion of the components, and a correction factor is applied according to the deviation magnitude. Adjustments are only performed when the deviation exceeds the threshold.

[0036] The total energy consumption is broken down step by step by dividing the system loops, and the component-level feature dataset is output in a structured form.

[0037] In a preferred embodiment, separating the independent energy consumption components of a specific component from the associated and matched energy consumption data includes:

[0038] Extract the energy consumption time series of the target component from the associated data;

[0039] Based on the changing characteristics of the time series, the energy consumption sequence is segmented, and the total energy consumption is divided into the startup phase, the stable operation phase, and the standby phase.

[0040] Energy consumption data is broken down by parameter type, and parameter type and stage labels are added to each data point;

[0041] Traverse the time series and classify data points into the corresponding stage and parameter type sets based on the labels;

[0042] Different energy sources within the same phase are broken down into energy consumption components, and a structured dataset is created.

[0043] In a preferred embodiment, the isolated independent energy consumption components are compared and analyzed against a predefined standard energy consumption model, including:

[0044] After separating the independent energy consumption components, they are compared and analyzed with the predefined standard energy consumption models for startup, operation, and standby in the database.

[0045] Align the data of each component with the model's time axis or load axis, calculate the difference between the actual and model power point by point, and record the magnitude and direction of the deviation;

[0046] In the comparative analysis, the actual operation phase is divided into load intervals and matched with the corresponding load points in the model to calculate the operational energy consumption deviation.

[0047] In the standby comparison, the actual power is compared with the model reference range to determine the power position;

[0048] The allowable consumption range is adjusted according to environmental conditions;

[0049] Output the separated data of the components, along with the corresponding model deviation indicators and a list of deviation points, in a structured format.

[0050] In a preferred embodiment, the energy consumption deviation value of a specific component is calculated using the independent energy consumption components after comparative analysis, and the calculated deviation value is integrated into the overall system diagnostic framework, including:

[0051] Read the deviation labels and matching indicators of each energy consumption component, calculate the percentage difference of actual power relative to the standard value, and generate a set of signed deviations.

[0052] The deviations are refined according to parameter type. The deviations of motor power and thermal loss power are calculated separately and labeled with type. When multiple components are coupled, an association label is added.

[0053] For the same component, the deviations are summarized by stage and the average is taken to generate a summary containing the component ID and the percentage of deviation in each stage, and the maximum deviation point is recorded;

[0054] Write the deviation values ​​into the diagnostic matrix, create rows and columns according to component and deviation type, add association tags, and output in file format.

[0055] In a preferred embodiment, the final output is an integrated diagnostic framework for continuous monitoring and adjustment of the central air conditioning system, including:

[0056] The diagnostic matrix is ​​formatted to generate a summary view of component deviations, and component and system deviations are calculated. When the deviations exceed a preset threshold, adjustment suggestions are given.

[0057] The diagnostic results are displayed through the interface and reports, presenting the stage deviations and textual suggestions of each component in a matrix format, and are automatically updated at preset intervals or triggered by events.

[0058] Control commands are generated based on the suggested items, the set values ​​are adjusted proportionally or in steps, and then sent to the device controller in combination with the parameter relationships.

[0059] After the adjustment is performed, the sensor is triggered to re-acquire parameters and update the diagnostic matrix. The feedback loop is executed at the configured frequency. When the deviation exceeds the threshold, an additional loop is triggered. The diagnostic matrix and loop status are maintained separately for each unit.

[0060] Compared with the prior art, the present invention provides a method for online monitoring and diagnosis of energy consumption of central air conditioning, which has the following beneficial effects:

[0061] 1. This invention, by deploying multiple types of sensors at key nodes of a central air conditioning system to collect real-time operating parameters such as temperature, pressure, flow rate, power, and speed, constructs thermodynamic, mechanical, and control deviation vectors through classification and benchmark comparison. Then, using deviation patterns and mapping tables, it identifies specific components exhibiting anomalies. Finally, it breaks down total energy consumption into independent energy consumption components by component and stage using time alignment and power allocation. After point-by-point comparison with a predefined standard energy consumption model, it calculates component-level energy consumption deviation values ​​for startup, operation, and standby phases. The deviation results are then written into a diagnostic matrix to drive closed-loop adjustments of operating parameters. Compared to traditional coarse assessment methods based on overall system energy consumption, this invention enables quantifiable, traceable, and precise assessment of energy waste in specific components, significantly improving the accuracy and location capability of quantifying energy waste at the component level in central air conditioning systems. This provides a reliable basis for energy-saving optimization and fault prevention, and solves the problem of low accuracy in quantifying energy waste in specific components of central air conditioning systems in traditional methods.

[0062] 2. This invention constructs an online closed-loop mechanism that integrates data acquisition, parameter classification, component identification, energy consumption separation, model comparison, deviation calculation, and parameter adjustment. During the long-term operation of the central air conditioning system, multi-dimensional operating parameters are continuously collected. The deviation baseline and diagnostic matrix are updated in a rolling manner within the central processing unit according to time windows, ensuring that each round of diagnosis is based on the latest operating conditions and involves component-level analysis. Combining standard energy consumption models and environmental compensation logic, energy consumption performance under different seasons and loads is dynamically corrected, preventing fixed experience thresholds from failing under load fluctuations. Simultaneously, based on the diagnostic results, control quantities such as compressor speed, condensing pressure, and fan and water pump flow rates are linked, creating a closed-loop iteration between energy consumption diagnosis and operational control. This improves the automation and response speed of strategy adjustments, reduces reliance on manual inspections and experience-based judgments, and enables stable and timely identification of energy consumption behaviors deviating from reasonable ranges even under complex operating conditions and long-term operation. This achieves dynamic management and control of the energy-saving potential and abnormal risks of the central air conditioning system. Attached Figure Description

[0063] Figure 1 This is a flowchart of a method for online monitoring and diagnosis of energy consumption of central air conditioning according to the present invention. Detailed Implementation

[0064] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0065] Example: Figure 1A method for online monitoring and diagnosis of central air conditioning energy consumption is presented, including:

[0066] S1: Collect various operating parameters of the central air conditioning system in real time and send the collected operating parameters to the central processing unit through the data transmission channel;

[0067] S2: The central processing unit performs a classification operation on the transmitted operating parameters and compares the classified operating parameters with the preset benchmark parameters one by one.

[0068] S3: Identify specific components in the central air conditioning system based on the deviation information obtained from the comparison, and associate and match the identified specific components with the corresponding energy consumption data;

[0069] S4: Isolate the independent energy consumption components of a specific component from the associated and matched energy consumption data, and compare and analyze the isolated independent energy consumption components with the predefined standard energy consumption model;

[0070] S5: Calculate the energy consumption deviation of a specific component using the independent energy consumption components after comparative analysis, and integrate the calculated deviation into the overall system diagnostic framework;

[0071] S6: The final output is an integrated diagnostic framework for continuous monitoring and adjustment of the central air conditioning system.

[0072] S1: Real-time acquisition of various operating parameters of the central air conditioning system, and transmission of the acquired operating parameters to the central processing unit via data transmission channel. Specifically:

[0073] Firstly, multiple sensors are deployed in key parts of the system. Temperature sensors are installed at the evaporator inlet and outlet, and the condenser inlet and outlet, ensuring the probes are tightly fitted to the heat exchanger tube walls or heat exchange plates, and secured with thermally conductive adhesive and mechanical fasteners to reduce contact thermal resistance. Pressure sensors are placed near the refrigerant suction and exhaust lines and important valves, connected to the pipelines via threaded interfaces, with sealing gaskets at the interfaces to reduce the risk of leakage. Flow sensors are installed on the pump outlet side of the chilled water and cooling water pipelines, as well as in the air passages of some fan coil units, secured with flange connections or insertion structures, and marked according to the fluid direction to ensure the actual flow direction is consistent with the sensor's measurement axis. Power sensors are connected to the compressor, circulating water pump, cooling tower fan, and indoor fan motors via clamp-on current transformers and voltage probes. The power supply line enables non-invasive measurement; the speed sensor is installed on the compressor shaft or near the coupling, as well as on the fan impeller shaft, and acquires speed signals through magnetic induction or optical reflection, and is maintained in a fixed relative position with the rotating element using a bracket; the temperature sensor can be a PT100 platinum resistance type, with an accuracy of approximately ±0.1℃; the pressure sensor is a piezoresistive type, with a range covering 0~5MPa and an accuracy of approximately ±0.5%FS; the flow sensor is a vortex shear or electromagnetic type, with an accuracy of approximately ±1%; the power sensor supports three-phase measurement, with an accuracy of approximately ±0.2%; the speed sensor can be a Hall effect type, with a resolution of 1rpm; all types of sensor housings and terminals meet a protection level of not less than IP65, and the wiring is protected by waterproof cable connectors and protective hoses to adapt to environments with condensate, dust, and vibration;

[0074] To ensure time consistency in data acquisition, the main controller acts as the time master station to synchronize the clocks of each sensor. The main controller periodically broadcasts time synchronization messages to each sensor or connects to an NTP time server to calibrate the local clock. Upon receiving the synchronization command, each sensor corrects its local timing unit to keep the sensor time deviation within a preset range. This preset range can be determined based on the sampling period and the time accuracy requirements of subsequent data analysis; for example, it can be set as a certain percentage of the sampling period, or a tolerance of no more than a few milliseconds based on historical operational data statistics. The main controller sets a fixed sampling frequency of at least once per second and sends sampling period configuration parameters via the bus. Each sensor's built-in timing unit performs sampling according to a unified cycle, thus controlling the evaporator... The system converts parameters such as condenser inlet and outlet temperatures, refrigerant suction and discharge pressures, chilled water and cooling water flow rates or air flow rates, instantaneous power of compressors and motors, and real-time speeds of compressors and fans into digital quantities. The sensors integrate an analog-to-digital converter to convert analog signals into digital data, adding a local timestamp and sensor number. The main controller can employ a polling mechanism or a priority-based scheduling mechanism during data acquisition. Priorities can be pre-configured based on the equipment's impact on energy consumption or its fault risk level, ensuring that data from critical equipment such as compressors and chilled water pumps are read preferentially within the same sampling period. Simultaneously, the main controller reads data from each sensor sequentially according to a pre-built system topology, ensuring that the data recorded at any given time corresponds to the actual physical flow direction.

[0075] The collected operating parameters are packaged by the front-end acquisition nodes and sent to the central processing unit through a preset data transmission channel. The transmission channel can be wireless, wired, or a combination of both. In wireless mode, the acquisition nodes can send data frames to the gateway node via Zigbee or Wi-Fi. Zigbee is suitable for low-power, short-range, multi-node networking scenarios, while Wi-Fi is suitable for data center environments with high bandwidth requirements. In wired mode, the acquisition nodes can communicate with the central controller via the Modbus RTU protocol on an RS485 bus, or upload data via Ethernet based on the TCP / IP protocol. The gateway device will then transmit the data via Zigbee... Data in e- or Wi-Fi networks is uniformly converted into Ethernet packets and then forwarded to the central processing unit to achieve convergence of heterogeneous networks. The data packet format includes a packet header, sensor identifier, parameter value field, timestamp field, and check field. The packet header records the sensor ID, parameter type code, and data length. Parameter values ​​are encoded using floating-point or fixed-point formats. The timestamp is based on a unified time base. The check field uses the CRC-16 algorithm to generate a checksum. For wireless transmission, the AES-128 algorithm can be used to encrypt the packets. For wired transmission, electromagnetic interference can be reduced by using shielded twisted-pair cables, proper grounding, and separating strong and weak current wiring.

[0076] The central processing unit (CPU) can be deployed on embedded servers or industrial control computers in a computer room, or in virtual computing instances in public or private cloud environments. When deployed locally, it can use a computing platform based on ARM or x86 architecture, configured with a multi-core processor and at least 4GB of memory, and run an operating system with real-time scheduling capabilities. When deployed in the cloud, the CPU establishes a connection with the field gateway through a secure API interface, transferring the collected data to cloud storage and computing resources. The CPU has redundant communication interfaces, including at least two network interfaces: one connected to a wired Ethernet channel and the other to a wireless or cellular network. Under normal circumstances, the primary link is used to transmit data. The CPU determines the status of the primary link through heartbeat detection or communication timeout counting. When multiple consecutive transmission failures or no response are received within a preset time, the link switching logic is triggered, automatically switching to the backup network interface. The preset time can be determined based on the sampling frequency and statistical results of network round-trip latency, for example, set to an integer multiple of multiple sampling periods to avoid misjudging short-term jitter while promptly detecting link failures.

[0077] On the data receiving side, the central processing unit recalculates the CRC-16 checksum for each arriving message and compares it with the checksum field carried in the message. If the checksum fails, an error log is recorded, and a retransmission request is sent to the front-end node. When the number of retransmissions exceeds a preset threshold, an alarm module is triggered. The retransmission threshold can be configured based on the network packet loss rate and the processing capacity of the front-end device, for example, selecting a retransmission limit of 3 to 5 times. If data cannot be successfully received even after exceeding this limit, an anomaly is determined. To prevent high-frequency collected data from arriving in a short period of time and causing buffer overflow, the central processing unit establishes a buffer queue based on the FIFO principle to reserve buffer space for data streams from each collection node. Each record in the queue retains the original parameter value, timestamp, sensor ID, and transmission channel identifier for subsequent processing according to time order and source. At the same time, the central processing unit is equipped with a local storage device, such as an SSD, to temporarily store operating parameters when the upstream network is interrupted or cloud services are unavailable. When the network connection with the external system is restored, the cached data is uploaded in batches according to time order to avoid data gaps during long-term monitoring.

[0078] In practical engineering deployments, the sensor layout is appropriately adjusted according to the structural characteristics of the central air conditioning system. When the system uses a screw compressor, an additional current sensor can be placed in the compressor starting circuit to collect the peak starting current. When the system uses a variable frequency fan, the PWM control signal or inverter output frequency related to speed control can be recorded based on the speed acquisition. To reduce the impact of mechanical vibration on measurement accuracy, vibration damping brackets or buffer pads can be equipped at the sensor installation points near the compressor and large fan. To improve reliability in remote buildings or distributed facility scenarios, the data transmission channel can be configured with wired Ethernet as the main channel and mobile communication network as the backup channel. The central processing unit can support building control protocols such as BACnet and output real-time operating parameters to the upper-level building automation system or energy management platform through a gateway or protocol conversion module. Through the above sensor deployment, sampling, and communication mechanisms, key operating parameters such as temperature, pressure, flow rate, power, and speed can be continuously and reliably uploaded to the central processing unit.

[0079] S2: The central processing unit performs a classification operation on the transmitted operating parameters, comparing each classified operating parameter with a preset baseline parameter. Specifically, this is implemented as follows:

[0080] In the central processing unit (CPU), the operating parameters uploaded via the data transmission channel are classified to organize the data flow in an orderly manner. The CPU reads each record from the receive buffer in chronological order, extracting the parameter identifier, parameter value, and timestamp. Subsequently, the parameters are classified according to their physical properties and their role in the central air conditioning system. Temperature parameters such as evaporator outlet temperature and condenser inlet temperature, as well as pressure parameters such as refrigerant suction pressure and discharge pressure, are classified into thermodynamic categories to characterize the heat exchange and pressure balance state in the refrigeration cycle. Flow parameters such as hot and cold water circulation flow rate and airflow rate, as well as power parameters such as compressor power consumption and fan motor power, are classified into... The parameters are categorized into mechanical types to characterize the delivery efficiency and energy input level of water and air systems; and control types, such as compressor speed and fan impeller speed, to characterize the adjustment status of the frequency converter or controller on the actuators. To achieve this classification, the central processing unit has a built-in rule engine and maintains a parameter mapping table. Each parameter identifier in the mapping table is pre-attached with a category label. During processing, the rule engine traverses the parameter list, queries the mapping table, and assigns a category, ensuring that the classification process is executed item by item without omission. When the system adds new monitoring quantities such as vibration and noise, the mapping table can be updated to classify the new parameters into the corresponding categories to adapt to different models and configurations of central air conditioning systems.

[0081] After the classification process is completed, the central processing unit immediately enters the comparison phase, comparing the classified operating parameters one by one with the preset benchmark parameters to establish a basis for system operating deviations. The benchmark parameters are derived from system design specifications and historical operating databases. For example, the standard operating temperature range can be set according to the manufacturer's manual and environmental standards, with the evaporator set at 5–15°C and the condenser at 30–45°C. The rated power benchmark for electrical equipment such as compressors is the nominal value under design load, derived from the equipment nameplate data. The standard speed benchmark is the controller's preset operating range, for example, the normal speed range of the fan is 1000–3000 rpm. The comparison process uses a step-by-step matching mechanism. First, thermodynamic parameters are compared, comparing the actual evaporator outlet temperature with the corresponding standard operating temperature range to determine if it falls within the allowable range and recording the direction of deviation. Then, pressure is compared... The system compares actual intake and exhaust pressures with their respective benchmark values ​​or ranges, calculates the offset, and marks the offset trend. Then, it moves to mechanical parameters, comparing actual hot and cold water flow rates and cooling water flow rates with design flow rates or historical calibration flow rates to check for long-term underestimation or short-term sudden changes. Power parameters are compared with actual power consumption and rated power or historical stable power, focusing on key time points of load increase, decrease, and fluctuation. Finally, control parameters are processed, comparing actual compressor and fan speeds with preset speed ranges to identify overspeed, underspeed, or frequent fluctuations. Through this item-by-item comparison mechanism, while ensuring that each parameter is evaluated independently, the central processing unit can also combine the physical relationships between parameters, such as adjusting the weight of pressure deviation when temperature deviations are significant, thereby forming a more comprehensive deviation basis that better reflects actual operating conditions.

[0082] When establishing the deviation basis, a unified measurement is achieved by constructing relative deviation values. For temperature parameters, the difference between the actual value and the median of the reference interval is compared with the width of the reference interval to obtain the normalized temperature deviation, reflecting the degree of deviation. For pressure parameters, the difference between the actual value and the reference value is expressed as a percentage of the reference value to obtain the pressure deviation. For flow parameters, the deviation of the ratio of the actual flow rate to the reference flow rate is used as the flow deviation. For power parameters, the ratio of the difference between the actual power and the rated power to the rated power is used as the power deviation. For speed parameters, a standardized speed deviation is constructed based on the difference between the actual speed and the center value of the preset speed interval, as well as the width of that interval. The central processing unit summarizes each of the above deviations to form a deviation vector organized by category. The thermodynamic deviation vector includes temperature and pressure deviation terms, the mechanical deviation vector includes flow and power deviation terms, and the control deviation vector includes speed deviation terms. The deviation basis constructed in this way provides clear information support for subsequent analysis, enabling the deviation patterns in the deviation vector to be used for component-level analysis, thereby avoiding the problems of scattered and unusable deviation information.

[0083] The design of this classification and comparison logic emphasizes modularity; the classification engine can be deployed as an independent sub-module and integrated with other functional modules of the central processing unit through a standard interface; the comparison mechanism can support parallel processing, allocating calculation threads for thermodynamic and mechanical parameters on multi-core processors to improve comparison efficiency; the preset benchmark parameters can be configured through the user interface, allowing maintenance personnel to enter the corresponding standard data for different air conditioner models, thus ensuring compatibility with both variable frequency and fixed frequency systems; simultaneously, the central processing unit can extract parameter statistics under stable operating conditions over a period of time from the historical operating database as a supplementary benchmark beyond the design specifications; the overall logic flow starts with data reception, proceeds through classification organization, and then to benchmark comparison and deviation-based output, resulting in a clear structure; the classification labels for sensor parameters can be defined and updated through XML configuration files, and key matching results generated during the comparison process can be written to log files for auditing and traceability, thereby facilitating the reproduction and expansion of this classification and comparison scheme in actual engineering projects by those skilled in the art;

[0084] In professional applications, it can be applied to multi-unit central air conditioning systems in large commercial buildings. For multi-unit scenarios, when performing classification operations, the central processing unit adds a unit identifier field to the original parameter categories, and performs dual indexing and storage of operating parameters according to unit ID and parameter category, so as to evaluate the operating conditions of each unit separately. The benchmark parameters used for comparison can incorporate environmental compensation logic. For example, the standard operating temperature range of evaporator and condenser can be dynamically adjusted according to environmental conditions such as outdoor temperature and humidity. When the outdoor temperature is high, the upper limit of condenser temperature can be appropriately relaxed, and when the ambient temperature is low, the lower limit of evaporator outlet temperature can be tightened. The output of the deviation basis can be encapsulated in a structured data format such as JSON, recording information such as unit ID, parameter category, normalized deviation value, and whether the preset deviation threshold is exceeded, and directly passing it to the subsequent fault identification and energy consumption assessment modules for calling. This ensures the continuity and scalability of the data flow from classification to deviation establishment to diagnostic analysis in different application scenarios.

[0085] S3: Identify specific components in the central air conditioning system based on the deviation information obtained from the comparison, and associate and match the identified specific components with the corresponding energy consumption data. Specifically, this is implemented as follows:

[0086] Based on the deviation vectors and configured deviation thresholds constructed within the central processing unit (CPU) as mentioned earlier, the CPU first reads the normalized deviation values, timestamps, parameter identifiers, and category information of each parameter in the thermodynamic, mechanical, and control deviation vectors from the deviation baseline. For each sampling moment or preset time window, the CPU performs aggregation analysis on the deviation vectors according to parameter category and position in the system topology, filtering out deviation points where the normalized deviation exceeds the preset deviation threshold. Only significant and representative deviation signals are retained to avoid minor fluctuations interfering with pattern recognition. The deviation threshold can be the normalized deviation value configured earlier. A standardized deviation threshold is established, for example, by determining the range of normal samples based on the allowable deviations in equipment design, relevant standards, and the statistical distribution of historical operating data. Subsequently, the central processing unit combines the deviation points exceeding the threshold into deviation mode units according to parameter type and location. For example, low temperature deviation is combined with the evaporator outlet location to form an evaporator-related mode, high pressure deviation is combined with the refrigerant circulation downstream location to form a condenser-related mode, peak power deviation is synchronously combined with speed deviation to form a compressor-related mode, and low flow deviation concentrated in a certain branch corresponds to a water pump or fan-related mode. This transforms single-point deviation information into a mode description with clear component orientation.

[0087] After a deviation pattern is formed, the central processing unit calls a preset deviation mapping table to complete component identification. The deviation mapping table is stored in a database or configuration file in the form of structured configuration data, which records the component identifier, its location region, typical deviation pattern combinations, and optional weight parameters. During system deployment, the system is first divided into components such as compressors, condensers, evaporators, fans, water pumps, and expansion valves according to the topology of the central air conditioning system. Several typical deviation patterns are preset for each component. For example, the compressor corresponds to a mode in which the power deviation reaches a peak and changes synchronously with the speed deviation; the condenser corresponds to a mode in which the pressure deviation is high and accompanied by temperature deviation in the condenser area; and the water pump corresponds to a mode in which the flow deviation is low and the pump is located in the middle section of the corresponding pipeline. During the identification phase, the central processing unit sequentially traverses each deviation pattern unit within the current time window and matches the pattern features as query conditions in the deviation mapping table. When multiple component candidates meet the same deviation pattern, they can be sorted by comprehensive indicators such as pattern weight, deviation amplitude, deviation duration, and number of matching patterns. The component identifier with the higher comprehensive weight is selected first. Finally, the deviation information is matched one-to-one with the specific component, forming a component identifier and its deviation pattern identification result recorded by time window.

[0088] After component identification is completed, the central processing unit (CPU) associates and matches the identified components with their corresponding energy consumption data to obtain component-level energy consumption characteristics. The energy consumption data originates from the raw operating parameters and their derived indicators collected and stored earlier, including real-time power measurement sequences, cumulative energy consumption, operating status markers, and start / stop records. The CPU uses the component identifier as the primary key to retrieve corresponding records in the energy consumption data table. For example, for the identified compressor component, it retrieves its three-phase power measurement data, operating status markers, and cumulative runtime; for the identified condenser component, it retrieves the fan or pump power data related to its heat exchange process and the corresponding operating range. During the association and matching process, the timestamp or time window in the deviation mode record is used as the time index. The time range of the deviation event is aligned with the time field in the energy consumption data. When the difference between the deviation time and the energy consumption recording time exceeds the preset alignment threshold, the energy consumption data is corrected by interpolation or expanding the time window to ensure the temporal consistency between the deviation and energy consumption. The alignment threshold can be determined based on the statistical results of the system sampling period, network transmission latency, and time synchronization accuracy. For example, it can be set to a multiple of the sampling period to avoid misjudging instantaneous jitter as time misalignment while maintaining alignment accuracy. After the association is completed, the central processing unit builds an initial association dataset for each component, which includes deviation information and energy consumption curves. For example, the association dataset for the compressor includes a normalized power deviation sequence, a power measurement sequence within the corresponding time period, and corresponding runtime information.

[0089] Based on the initial associated dataset, the central processing unit further splits and isolates the energy consumption data to extract relatively independent energy consumption characteristics of specific components. For total power measurements involving multiple components in the same time period, a baseline allocation of total power can be made based on the operating status, rated power percentage, or historical stable operating condition power percentage of each component during that time period. The allocation coefficient is then adjusted according to the normalization magnitude of the current deviation, so that components with larger deviations receive a relatively higher energy consumption allocation ratio during that time period. To avoid frequent fluctuations in energy consumption allocation due to small deviations, the adjustment of the allocation coefficient can only be triggered when the corresponding normalized deviation exceeds the aforementioned deviation threshold. For scenarios with series or parallel system structures, the loop delineation in the system drawings can be used as a reference. The total energy consumption is broken down step by step according to the loop and component levels, ultimately obtaining the isolated power sequence and corresponding deviation sequence for each component. After the breakdown is completed, the central processing unit outputs the component-level features in a structured form, such as using tables or JSON arrays to record component IDs, timestamps, deviation labels, and isolated power or energy consumption values, and passes this feature dataset to the subsequent energy consumption assessment and fault diagnosis modules. Through the processing flow from deviation basic input, pattern recognition, component mapping to energy consumption association and feature isolation, the operating deviations and energy consumption performance of each key component in the central air conditioning system can be output in the form of component-level quantitative results under a unified framework, and are closely connected with the data acquisition and transmission process and deviation basic construction process mentioned above.

[0090] S4: Isolate the independent energy consumption components of specific parts from the associated and matched energy consumption data, and compare and analyze the isolated independent energy consumption components with the predefined standard energy consumption model. The specific implementation is as follows:

[0091] The central processing unit (CPU) separates the independent energy consumption components of specific components from the associated energy consumption data and compares these components with a predefined standard energy consumption model. Based on the aforementioned component-level energy consumption characteristics, the CPU extracts the energy consumption time series of the target component from the associated data. The energy consumption time series is stored in chronological order, and each data point includes at least a power value, a timestamp, and an operating status or type label. For example, for a compressor, the energy consumption time series includes power measurement points and corresponding operating status labels throughout the entire process from startup to stable operation. The CPU segments the time series according to its variation characteristics, dividing the total energy consumption sequence of the component into sub-intervals such as the startup phase, stable operation phase, and standby phase. The startup phase is defined as the time interval during which power rapidly rises from the baseline value to near the steady-state value. The operation phase is defined as a continuous time interval during which power and speed are basically stable. The standby phase is defined as an intermittent interval during which power is lower than a preset standby power threshold. The power change rate threshold and the standby power threshold can be determined based on the startup characteristic curve, rated power, and statistical results under historical stable operating conditions in the equipment manual, and configured in the system in the form of a threshold table to facilitate adjustments for different equipment models.

[0092] After completing the segmentation, the central processing unit further subdivides the energy consumption data within each segment according to parameter type. During the data acquisition or preprocessing stage, type and stage labels are pre-added to each energy consumption data point. The stage label identifies the startup, operation, and standby stages, while the type label identifies information such as motor drive power, mechanical loss power, thermal loss power, and standby duration. During separation, the central processing unit traverses the energy consumption time series of the target component. First, it divides the data points into different stages such as startup, operation, and standby based on the stage label. Then, it breaks down different energy consumption sources within the same stage into several components based on the type label. For example, in the startup stage, the power integral generated during motor acceleration is taken as the startup energy consumption component; in the operation stage, the steady-state... The power integral under load is considered as a component of operating energy consumption. During the standby phase, the power integral required for maintaining control and monitoring is considered as a component of standby energy consumption. A label-based hierarchical classification method ensures that the same data point belongs to only one energy consumption component, avoiding overlap between different components. After separation, the central processing unit establishes structured datasets for components such as startup, operation, and standby. The startup energy consumption dataset includes time series points, power values, phase labels, type labels, and the total energy consumption integral of the startup phase. The operation energy consumption dataset includes the power curve, peak and valley values, and average power of the stable operation range. The standby energy consumption dataset includes the standby time period and the corresponding baseline power, thereby obtaining multi-dimensional energy consumption composition information for individual components.

[0093] After obtaining the individual energy consumption components, the central processing unit (CPU) compares these components with predefined standard energy consumption models. These standard energy consumption models are stored in a database as component-specific reference libraries. For compressors, the predefined standard energy consumption models can include startup energy consumption models, operating energy consumption models, and standby energy consumption models. The startup energy consumption model provides the power-time curve during an ideal startup process; the operating energy consumption model provides the expected energy consumption or efficiency range with load rate as the independent variable; and the standby energy consumption model provides the typical standby power range under different operating conditions. These standard energy consumption models can be constructed based on nominal parameters provided by the equipment manufacturer, bench test data, and simulation analysis results, and are stored in curve or lookup table form. During the comparative analysis, the CPU first aligns the data of each energy consumption component with the time axis or load axis of the corresponding standard model. In the startup energy consumption comparison, the time axis of the separated startup phase is aligned with the time axis of the standard startup curve. Then, the difference between the actual power and the model power is calculated point by point, and the deviation value and direction are recorded. In the operation energy consumption comparison, the actual operation phase is divided according to the load interval and matched with the corresponding load point in the operation energy consumption model. The energy consumption deviation or efficiency deviation of each load point is calculated. In the standby energy consumption comparison, the actual power of each standby time period is compared with the reference range given by the standby model to determine whether the power is consistently higher than the upper limit of the reference or lower than the lower limit of the reference. During the comparison process, the central processing unit expresses the deviation results in both absolute value and relative percentage forms, and calculates matching indicators such as the mean deviation, maximum deviation, and number of deviation points for each energy consumption component, thereby forming quantitative results such as the matching degree of the startup phase, the matching degree of the operation phase, and the matching degree of the standby phase.

[0094] To adapt to different environmental conditions, the standard energy consumption model can support environmental parameter compensation. For example, the operating energy consumption model can be corrected based on outdoor temperature or cooling water inlet temperature. When the ambient temperature is high, the allowable consumption range of the model can be appropriately increased, and when the ambient temperature is low, the range can be tightened, so that the comparison results are more consistent with the actual operating conditions. After comparison analysis, the central processing unit outputs the separated data of each energy consumption component, along with the corresponding model deviation index and deviation point annotations, in a structured form. For example, it records the component ID, component type, time range, matching degree index, and deviation point list as a JSON object, and passes these results to the subsequent energy consumption deviation assessment and diagnostic decision-making module. Through the processing flow from component-level energy consumption feature input, time series segmentation and type splitting, standard energy consumption model comparison to deviation index output, the energy consumption behavior of key components of the central air conditioning system in different stages such as startup, operation, and standby can be refined, decomposed, and quantitatively characterized, maintaining overall linkage with the data acquisition, deviation construction, and component identification processes mentioned above.

[0095] S5: Calculate the energy consumption deviation of specific components using the independent energy consumption components after comparative analysis, and integrate the calculated deviation values ​​into the overall system diagnostic framework. Specifically, this is implemented as follows:

[0096] The energy consumption deviation of specific components is calculated using the independent energy consumption components obtained from the comparative analysis. The calculated deviation values ​​are then integrated into the overall system diagnostic framework. Based on the independent energy consumption components obtained through separation and comparative analysis mentioned earlier, the central processing unit directly reads the deviation point annotations and matching degree indicators carried by each component and performs deviation calculations for each component separately. For components like compressors, the central processing unit calculates the energy consumption deviation as a percentage for the start-up energy consumption components, operating energy consumption components, and standby energy consumption components, using the difference between the recorded actual power curves or average power values ​​and the corresponding values ​​in the standard energy consumption model. The calculation method for operating energy consumption deviation is as follows: first, the actual average power during the operating phase is compared with the corresponding load point in the standard model. The expected power is subtracted to obtain the absolute power difference. This difference is then divided by the standard power and converted to a percentage to obtain the operating deviation. The start-up energy consumption deviation is obtained by comparing the energy integral under the actual start-up curve with the energy integral under the standard start-up curve, calculating the ratio of the difference to the standard integral, and obtaining the start-up deviation percentage. The standby energy consumption deviation is obtained by comparing the average actual standby power with the standard standby power constant or the median of the reference range. Through these calculations, different components such as start-up, operation, and standby are assigned deviation values ​​with positive and negative signs. Positive deviations indicate energy consumption higher than the standard, and negative deviations indicate energy consumption lower than the standard, thus forming a set of deviation values ​​divided by component. For example, a compressor might have a start-up deviation of +12%, an operation deviation of +8%, and a standby deviation of +5%.

[0097] After the set of deviation values ​​is generated, the central processing unit further refines the deviation results according to parameter type to distinguish the contribution of different energy consumption sources. For example, within the operating energy consumption component, the motor drive power deviation and heat loss power deviation are calculated separately and marked as mechanical power deviation and heat loss deviation by type sub-labels to avoid simple mixing during aggregation. For the fan component, the power deviation related to air flow and the power deviation related to speed control can be calculated separately and synthesized into operating deviation according to preset weights when a component is aggregated. For the water pump component, the power deviation related to the initial flow can be calculated separately during the startup phase. The central processing unit traverses all independent energy consumption components in chronological order, completes the calculation of deviations in the startup, operation, and standby phases in sequence, and adds labels such as component ID, component type, parameter type, and deviation symbol to each deviation value. In multi-component coupling scenarios, if the analysis results show that the deviation of a certain component is affected by upstream or downstream components, such as the correlation between compressor operating deviation and evaporator heat transfer deviation, a correlation label is added to the deviation record for subsequent diagnosis to interpret and correct the deviation baseline, avoiding misjudgment caused by interpreting only from the perspective of a single component.

[0098] After calculating the deviations for each component, the central processing unit (CPU) summarizes the deviation values ​​to construct a component-level deviation summary. The summarization process begins with the deviation set for the same component, aggregating the deviation percentages for the component during startup, operation, and standby phases. For example, it averages or weighted averages the deviation percentages across multiple time windows during the operation phase to obtain the representative deviation for that component during operation. The same method is used for the startup and standby phases to obtain startup and standby phase deviations, respectively. Subsequently, the CPU further groups the deviations according to component type and parameter type based on the deviation labels. For example, power-related deviations are categorized into mechanical types, and speed-related deviations into control types. When necessary, the CPU calculates the overall deviation for mechanical and control types separately. The summarization process also extracts extreme value information, such as recording the maximum deviation point and its time location during operation to highlight critical anomalies. Finally, the CPU generates a structured deviation summary record for each component, including the component ID, startup phase deviation percentage, operation phase deviation percentage, standby phase deviation percentage, overall deviation percentage, and a description of the maximum deviation, thus forming a traceable and auditable deviation summary list.

[0099] After obtaining the component-level deviation summaries, the central processing unit integrates these deviation values ​​into the overall system diagnostic framework to form a unified diagnostic basis. The diagnostic framework can be designed as a matrix or table structure, where rows correspond to various components in the system, such as compressors, condensers, evaporators, fans, and water pumps, and columns correspond to different deviation types, such as start-up deviations, operating deviations, standby deviations, overall deviations, and optional mechanical and control deviations. The integration process begins by initializing the diagnostic matrix, creating matrix rows based on the component IDs in the component library, creating matrix columns based on predefined deviation types, and then filling the list row by row according to the deviation summary list, adding the start-up deviation values ​​of each component. Dynamic deviation, operational deviation, and standby deviation are written into the corresponding cells. When a component has an additional deviation type, such as an airflow deviation for a fan, the matrix can be expanded to add new columns as needed. For component combinations with strong coupling relationships, the central processing unit can also add association marker columns or additional description fields to the matrix to record information such as the linkage between compressor operational deviation and fan operational deviation, facilitating joint analysis during subsequent diagnosis. To ensure that the diagnostic framework covers all target components, the central processing unit verifies against the component library after filling in the data. If a component is found to lack a deviation summary, it will be processed by marking the status of that row or triggering supplementary calculations.

[0100] After integration, the diagnostic framework can serve as the foundational data structure for system-level diagnostics, outputting data via files or interfaces, such as exporting matrix data in XML, CSV, or JSON formats. This matrix data contains all components, deviation types, and related annotation information. Subsequent diagnostic, energy consumption optimization, or operation and maintenance decision-making modules can directly read this matrix and generate alarms, prioritize maintenance, or provide parameter adjustment schemes based on deviation levels and component importance. Through a processing flow from inputting independent energy consumption components, deviation calculation and refinement, component-level deviation summary construction, to the integrated output of the diagnostic matrix, the energy consumption deviations of key components in the central air conditioning system can be quantified and compared within a unified framework. Furthermore, this diagnostic framework forms a complete analysis chain with the data acquisition, deviation construction, component identification, and energy consumption separation and comparative analysis processes mentioned earlier.

[0101] S6: The final output is an integrated diagnostic framework for continuous monitoring and adjustment of the central air conditioning system, specifically implemented as follows:

[0102] The final output is an integrated diagnostic framework for continuous monitoring and adjustment of the central air conditioning system. Based on the unified diagnostic criteria formed through deviation calculation, summarization, and matrix integration mentioned above, the central processing unit formats the existing diagnostic matrix, transforming the internal multidimensional data structure into an application-oriented output framework. Following a preset output template, the central processing unit maps the matrix rows and columns into a hierarchical list or table structure, first listing the component name, then listing the deviation values ​​and corresponding descriptions for that component during startup, operation, and standby phases. For example, when the operating deviation is positive and large, a prompt to reduce the load or optimize cooling conditions is given. Based on this, the central processing unit generates a deviation summary view, such as calculating the overall deviation percentage by component and the average deviation level by system, and generating corresponding prompts based on the deviation direction and magnitude. When a deviation index exceeds a preset deviation threshold, corresponding adjustment suggestions are automatically added to the output framework. The deviation threshold and suggestion rules can be configured and updated by maintenance personnel in conjunction with energy-saving management strategies, equipment manual parameters, and historical operating statistics.

[0103] The output framework can be presented through either a user interface or offline reports. The user interface can be a graphical monitoring page, displaying diagnostic results on a host computer, webpage, or mobile terminal. For example, the main interface can display a matrix view or device topology diagram arranged by components. Selecting a component expands the graph with deviation bars or line charts for the component during startup, operation, and standby phases, displaying corresponding text suggestions and alarm indicators below the graph. The report format is achieved by generating document files, such as PDFs or spreadsheets. The homepage provides a system-level deviation summary table, listing the percentage deviation of each component at each stage and the overall deviation. Subsequent pages list detailed deviation curves, deviation point descriptions, and suggested items for each component, indicating the reference standard or model source for the deviation in appropriate locations. The central processing unit can update the user interface and report content according to preset time intervals or event trigger conditions, such as automatically refreshing the interface data when a new diagnostic matrix is ​​generated, or pushing alarm information to designated users when the deviation exceeds a preset deviation threshold. The system can also support user views with different permissions; for example, the administrator view can view the complete deviation matrix and historical records, while the ordinary operation and maintenance view only displays summary information and actionable operation suggestions, adapting to the needs of different roles.

[0104] Based on the output diagnostic framework, the central processing unit (CPU) adjusts the parameters of the central air conditioning system according to the suggested items to achieve closed-loop control from diagnosis to optimization. The CPU reads the adjustment items to be executed from the diagnostic framework. For example, if the compressor operating deviation is positive and exceeds the aforementioned preset deviation threshold, it generates control commands to reduce the compressor speed or load rate; if the condenser-related deviation is positive, it generates control commands to lower the condensing pressure setting or increase the cooling water flow rate. The adjustment range can be determined according to a preset proportion or step size, such as increasing or decreasing by a certain percentage relative to the current setpoint, or switching between multiple operating levels according to preset gears, while considering the equipment's permissible operating range and safety considerations. The central processing unit (CPU) adjusts the rotational speed of components such as fans and pumps according to the deviation of air or water flow to restore the target operating conditions. During adjustments, the CPU comprehensively considers the relationships between various operating parameters. For example, while reducing compressor speed, it monitors fan operating status and condensing pressure changes to avoid new temperature or pressure anomalies caused by interlocking changes. Control commands are sent to field device controllers via preset communication protocols, such as sending speed settings to the inverter via Modbus and valve opening or pressure settings to the actuator via building automation bus, to ensure that adjustments are completed within an acceptable timeframe and to record corresponding command logs in the control system.

[0105] To achieve continuous monitoring and dynamic adjustment, a feedback loop is introduced between diagnosis and adjustment, using the adjusted system state as input for the next round of analysis. After executing a set of parameter adjustment operations, the central processing unit triggers sensors to re-collect key operating parameters, including temperature, pressure, flow rate, power, and speed. The newly collected data is sent back to the central processing unit through the aforementioned data transmission channel and sequentially undergoes the aforementioned data classification and comparison, component identification and energy consumption isolation, energy consumption component separation and standard model comparison, and deviation calculation and matrix integration processes to generate an updated diagnostic matrix. The execution frequency of the feedback loop can be configured according to system scale, load fluctuations, and energy-saving strategies. For example, under normal operating conditions, a cycle of minutes or ten minutes can be used, while the cycle can be appropriately shortened when environmental conditions change rapidly or deviations fluctuate significantly. Additionally, an extra cycle can be triggered immediately when a deviation change exceeds a preset threshold. This threshold can be configured based on historical deviation fluctuation statistics, system responsiveness, and operational strategies. At the end of each cycle, the central processing unit compares the changes in deviation indicators before and after the adjustment. If it finds that an adjustment has failed to effectively reduce the deviation or has even increased it, the corresponding parameter adjustments can be rolled back according to the configured strategy, and relevant logs can be recorded so that operations personnel can analyze the causes and optimize subsequent adjustment rules.

[0106] The feedback loop design also supports multi-unit and multi-region application scenarios. The central processing unit can maintain the diagnostic matrix and loop status separately for each unit or region, allowing different units to execute monitoring and adjustment processes independently in logic. When the compressor deviation of a certain unit reaches the adjustment condition, parameter adjustment and feedback loop are only executed for that unit without affecting the normal operation strategy of other units. Deviation summaries and adjustment records are continuously recorded during operation, forming a deviation trend and adjustment history arranged on a time axis, which can be used for long-term statistical analysis and strategy optimization. The output diagnostic framework serves as the starting point for parameter adjustment and feedback loop. On the one hand, the deviations and suggestions in the framework drive the adjustment of the operating parameters of the field equipment. On the other hand, the new data after adjustment updates the diagnostic matrix through the aforementioned data processing link, thus forming a closed-loop process from unified diagnostic basis input, framework formatted output, parameter optimization adjustment to feedback re-diagnosis. This enables the central air conditioning system to continuously achieve dynamic monitoring and optimization of energy consumption and operating conditions during the operating cycle, and facilitates the direct deployment and expansion of this implementation method in different engineering projects by those skilled in the art.

[0107] In this embodiment, multiple sensors are first deployed at key nodes of the central air conditioning system. Temperature sensors are placed at the inlet and outlet of the evaporator and condenser, pressure sensors are placed in the refrigerant pipeline, flow sensors are embedded in the water pump and fan pipeline, power sensors are connected to the compressor and motor, and speed sensors are installed at the relevant shaft ends. Operating parameters such as temperature, pressure, flow, power, and speed are collected at a frequency of at least once per second and transmitted to the central processing unit via wired or wireless channels such as Modbus and Ethernet. In the central processing unit, the collected parameters are classified into thermodynamic, mechanical, and control categories, and compared with preset benchmarks one by one to build a deviation basis. Based on the deviation information and the deviation mapping table, specific components such as compressors and condensers are identified, and they are associated with corresponding energy consumption data and isolated from component-level energy consumption characteristics. Subsequently, independent energy consumption components such as compressor start-up, operation, and standby are separated according to time series and parameter type, and compared with standard energy consumption models one by one to calculate the percentage difference in energy consumption at each stage. The results are summarized to form component-level energy consumption deviations and written into the diagnostic matrix. The deviations and energy-saving suggestions are displayed through the interface or report, driving the adjustment of operating parameters such as compressor speed and condensing pressure. The adjusted new data re-enters the acquisition and analysis process, forming a closed-loop dynamic optimization control for central air conditioning energy consumption.

[0108] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0109] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wireless or wired transmission; wired transmission methods include optical fiber, twisted pair, coaxial cable, etc.; wireless transmission includes infrared, microwave, etc. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0110] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0111] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0112] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0113] In addition, the functional modules in the embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0114] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0116] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for online monitoring and diagnosis of energy consumption in central air conditioning systems, characterized in that, include: S1: Collect various operating parameters of the central air conditioning system in real time and send the collected operating parameters to the central processing unit through the data transmission channel; S2: In the central processing unit, the transmitted operating parameters are classified. The classified operating parameters are compared with the preset benchmark parameters one by one. Records are read from the receiving buffer in chronological order, and the parameter identifier, value and timestamp are extracted. The parameters are classified according to their physical attributes and their role in the system. Category labels are bound to each parameter. When a new monitoring quantity is added, the mapping table is updated to complete the classification. The classified operating parameters are compared with the preset benchmark parameters one by one to construct a unified measure of relative deviation. The actual values ​​of each parameter are normalized and compared with their respective benchmark values. The deviations are summarized to form a deviation vector organized by category. S3: Identify specific components in the central air conditioning system based on the deviation information obtained from the comparison. Read the normalized deviation values, timestamps, parameter identifiers, and category information of thermodynamic, mechanical, and control deviation vectors from the deviation basis. Aggregate deviation vectors according to parameter category and system topology location. Filter deviation points where the normalized deviation exceeds the preset threshold and combine them into deviation pattern units. Call the preset deviation mapping table to complete component identification. Store component identifiers, location areas, and typical deviation pattern combinations. Traverse the deviation patterns in the current time window, match features in the deviation mapping table, and select component identifiers according to pattern weight, deviation amplitude, and duration. Associate and match the identified specific components with the corresponding energy consumption data. After component identification is completed, associate and match the identified components with the corresponding energy consumption data. Align energy consumption time with the deviation timestamp as the index. When the alignment threshold is exceeded, correct the energy consumption data through interpolation or extended window. Build an associated dataset containing deviation information and energy consumption curves for each component. Allocate the total power according to the component's operating status and rated proportion, and adjust the coefficient according to the deviation amplitude. Adjustment is only performed when the deviation exceeds the threshold. Combine the system loop division to decompose the total energy consumption level by level and output the component-level feature dataset in a structured form. S4: Isolate the independent energy consumption components of a specific component from the associated and matched energy consumption data, and compare and analyze the isolated independent energy consumption components with the predefined standard energy consumption model; S5: Calculate the energy consumption deviation value of a specific component using the independent energy consumption components after comparative analysis, integrate the calculated deviation value into the overall system diagnostic framework, read the deviation label and matching index of each energy consumption component, calculate the percentage difference of actual power relative to the standard value, generate a set of signed deviations, refine the deviations according to parameter type, calculate the deviations of motor power and heat loss power separately and add type labels, add association labels when multiple components are coupled, summarize the deviations of the same component by stage and take the average, generate a summary containing component ID and stage deviation percentage, record the maximum deviation point, write the deviation value into the diagnostic matrix, establish rows and columns according to component and deviation type, supplement association tags and output in file form; S6: The final output is an integrated diagnostic framework that continuously monitors and adjusts the central air conditioning system. It formats the diagnostic matrix, generates a summary view of component deviations, calculates component and system deviations, and provides adjustment suggestions when deviations exceed preset thresholds. The diagnostic results are displayed through the interface and reports, presenting the stage deviations and text suggestions of each component in a matrix format. The system updates automatically at preset intervals or triggered by events. It generates control commands based on the suggested items, adjusts the setpoints proportionally or in steps, and sends them to the equipment controller in conjunction with parameter relationships. After the adjustment is executed, the sensors are triggered to re-collect parameters and update the diagnostic matrix. The feedback loop is executed at the configured frequency. When the deviation exceeds the threshold, an additional loop is triggered. The diagnostic matrix and loop status are maintained separately for each unit.

2. The method for online monitoring and diagnosis of central air conditioning energy consumption according to claim 1, characterized in that, The system collects various operating parameters of the central air conditioning system in real time and sends these parameters to the central processing unit via a data transmission channel. These parameters include: Temperature, pressure, flow, power and speed sensors are placed in key locations, using adhesive, threaded, flanged or insert, clamp-on current transformers and magneto-optical mounting. The clock of each sensor is synchronized, time synchronization messages are broadcast periodically, and configuration parameters with a fixed sampling frequency are sent out. Data is read through polling or priority scheduling, and parameters are recorded in the order of system topology. The front-end node packages the data and sends it to the central processing unit via wireless or wired channels, employing encryption algorithms and electromagnetic interference protection measures. The receiving side calculates the checksum and processes messages that fail the checksum verification; at the same time, it temporarily stores the data in a buffer queue and uploads it in batches after the network is restored.

3. The method for online monitoring and diagnosis of central air conditioning energy consumption according to claim 1, characterized in that, Isolate the independent energy consumption components of a specific component from the associated and matched energy consumption data, including: Extract the energy consumption time series of the target component from the associated data; Based on the changing characteristics of the time series, the energy consumption sequence is segmented, and the total energy consumption is divided into the startup phase, the stable operation phase, and the standby phase. Energy consumption data is broken down by parameter type, and parameter type and stage labels are added to each data point; Traverse the time series and classify data points into the corresponding stage and parameter type sets based on the labels; Different energy sources within the same phase are broken down into energy consumption components, and a structured dataset is created.

4. The method for online monitoring and diagnosis of central air conditioning energy consumption according to claim 1, characterized in that, The isolated independent energy consumption components are compared and analyzed with predefined standard energy consumption models, including: After separating the independent energy consumption components, they are compared and analyzed with the predefined standard energy consumption models for startup, operation, and standby in the database. Align the data of each component with the model's time axis or load axis, calculate the difference between the actual and model power point by point, and record the magnitude and direction of the deviation; In the comparative analysis, the actual operation phase is divided into load intervals and matched with the corresponding load points in the model to calculate the operational energy consumption deviation. In the standby comparison, the actual power is compared with the model reference range to determine the power position; The allowable consumption range is adjusted according to environmental conditions; Output the separated data of the components, along with the corresponding model deviation indicators and a list of deviation points, in a structured format.

Images