An air conditioner control system based on flat design

Abstract

This invention discloses an air conditioning control system based on a flat design, specifically relating to the field of energy-saving air conditioning system safety monitoring technology. It addresses the problem of difficulty in timely identification and handling of hidden performance degradation in existing flat-architecture air conditioning systems. The system uses a parameter acquisition module to collect real-time equipment operating parameters and environmental parameters, an energy efficiency analysis module to calculate energy efficiency deviation data, and a step-out analysis module to analyze the step-out rate of the equipment's collaborative operation energy consumption trajectory. When energy efficiency is abnormal and the step-out rate continues to increase, a joint analysis module further uses Granger causality analysis and power spectrum asymmetry index calculation. A performance judgment module accurately determines whether the system has hidden performance degradation. The result execution module automatically generates equipment maintenance instructions and adjusts control strategies based on the judgment results, achieving early detection and adaptive regulation of system performance degradation, effectively ensuring stable operation of the air conditioning system in an energy-saving state.

Description

Technical Field

[0001] This invention relates to the field of safety monitoring technology for energy-saving air conditioning systems, and in particular to an air conditioning control system based on a flat design. Background Technology

[0002] In industrial buildings and large facilities, energy management of centralized air conditioning systems is a critical issue. To improve control efficiency and reduce energy consumption, air conditioning systems are increasingly adopting flat network architectures to replace traditional distributed control modes. These architectures use a centralized controller to directly connect and manage all field devices, aiming to reduce intermediate steps, improve response speed, and achieve global energy-saving optimization. In existing technologies, air conditioning systems based on flat architectures rely on optimization algorithms from a central controller to dynamically adjust equipment operating states based on real-time environmental parameters to achieve the desired temperature and humidity control and energy efficiency targets.

[0003] However, existing flat-control air conditioning systems have the following problems: when the system experiences hidden performance degradation such as filter clogging or scale buildup on heat exchange surfaces, the central controller will automatically increase the equipment output to compensate for the performance loss in order to maintain stable environmental parameters. This process masks the existence of the fault, and the system therefore operates in a low-energy-efficiency state for a long time without being detected in time, leading to increased energy waste. Moreover, the fault is difficult to detect through conventional monitoring mechanisms until the energy efficiency drops significantly or the equipment is overloaded, affecting the system's safety monitoring and energy-saving operation objectives. Summary of the Invention

[0004] This invention addresses the technical problems existing in the prior art by providing an air conditioning control system based on a flat design.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] This invention provides the following technical solution:

[0007] An air conditioning control system based on a flat design includes:

[0008] The parameter acquisition module is used to collect the operating parameters and corresponding environmental parameters of each device in the system in real time.

[0009] The energy efficiency analysis module is used to calculate the current actual energy efficiency value based on operating parameters and environmental parameters, and compare it with the preset benchmark energy efficiency value to generate energy efficiency deviation data;

[0010] The step loss analysis module is used to identify the period during which the system is in a stable temperature state. It analyzes the step loss rate of energy consumption trajectory by extracting the energy consumption data of multiple devices operating collaboratively during the corresponding maintenance period.

[0011] The joint analysis module is used to analyze the Granger causal relationship between the total energy consumption time series and the key control command time series when the energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, and to calculate the asymmetry index of the instantaneous power spectrum of the system.

[0012] The performance determination module is used to determine that if the Granger causality strength is less than a preset strength threshold and the asymmetry index continues to increase synchronously, the system has hidden performance degradation.

[0013] The result execution module is used to generate equipment maintenance instructions and adjust control strategies based on the results of the implicit performance degradation determination.

[0014] Furthermore, the system collects real-time operating parameters and corresponding environmental parameters of each device, including:

[0015] The current, power, and operating frequency data of the compressor, fan, and water pump in the air conditioning system are periodically and synchronously collected as operating parameters.

[0016] Real-time temperature and humidity data of the air conditioning system service area are collected synchronously as environmental parameters.

[0017] Furthermore, the current actual energy efficiency value is calculated based on operating parameters and environmental parameters, and compared with the preset benchmark energy efficiency value to generate energy efficiency deviation data, including:

[0018] Based on real-time collected operating and environmental parameters, the ratio of the total cooling capacity to the total power consumption of the air conditioning system per unit time is calculated as the current actual energy efficiency value.

[0019] The current actual energy efficiency value is compared with the energy efficiency benchmark value obtained from the system's historical healthy operating conditions, and the percentage deviation between the actual energy efficiency value and the preset benchmark energy efficiency value is calculated to generate energy efficiency deviation data.

[0020] Furthermore, the system identifies periods during which it maintains a stable temperature state. By extracting energy consumption data from multiple devices operating collaboratively within these periods, the system analyzes the energy consumption trajectory synchronization rate, including:

[0021] When the real-time temperature of the area served by the air conditioning system remains within the preset temperature fluctuation range and exceeds the preset duration, the system is determined to enter a period of temperature stabilization.

[0022] Extract the power consumption data of the compressor and fan operating in coordination during the maintenance period, and generate compressor power time series and fan power time series respectively;

[0023] Calculate the dynamic time warping distance between the compressor power time series and the fan power time series, and use the dynamic time warping distance as the energy consumption trajectory step loss rate.

[0024] Furthermore, calculating the dynamic time-warped distance between the compressor power time series and the fan power time series includes: using a dynamic programming algorithm to calculate the Euclidean distance between all data point pairs between the compressor power time series and the fan power time series and constructing a cumulative distance matrix; finding the optimal path from the upper left corner to the lower right corner in the cumulative distance matrix and using the minimum cumulative distance value on the optimal path as the energy consumption trajectory step loss rate.

[0025] Furthermore, when the energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, the Granger causality strength between the system's total energy consumption time series and the key control command time series is analyzed, and the asymmetry index of the system's instantaneous power spectrum is calculated, including:

[0026] When the energy efficiency deviation data exceeds the preset deviation threshold for multiple consecutive sampling periods and the energy consumption trajectory out-of-step rate shows a monotonically increasing trend, a system total energy consumption time series is constructed based on the system total power consumption data during the maintenance period, and a key control command time series is constructed based on the frequency control commands issued by the central controller to the compressor during the maintenance period. A Granger causality test is performed on the system total energy consumption time series and the key control command time series, and the F statistic value in the test result is used as the strength of the Granger causality relationship.

[0027] The power spectrum is obtained by performing a fast Fourier transform on the total power consumption data of the system during the maintenance period. The skewness of the corresponding power spectrum distribution relative to its center frequency is calculated as an index of the asymmetry of the instantaneous power spectrum of the system.

[0028] Furthermore, the Granger causality test for the total system energy consumption time series and the critical control command time series includes: using past values ​​of the critical control command time series to predict the current value of the total system energy consumption time series, while not using past values ​​of the critical control command time series to predict the current value of the total system energy consumption time series; comparing the error degree of the prediction results of the total system energy consumption time series under the two prediction methods; and calculating the F statistic value as the Granger causality strength based on the difference in error degree.

[0029] Furthermore, the calculation of the skewness of the power spectrum distribution corresponding to the total power consumption data of the system during the maintenance period relative to its center frequency includes: performing a fast Fourier transform on the total power consumption data of the system to obtain the power spectrum, calculating the center frequency of the power spectrum, and then calculating the ratio of the third central moment of the power spectrum distribution to the cube of the standard deviation, using the ratio as an indicator of the asymmetry of the instantaneous power spectrum of the system.

[0030] Furthermore, if the Granger causality strength is less than a preset strength threshold and the asymmetry index increases synchronously and continuously, then the system is determined to have implicit performance degradation, including:

[0031] Compare the Granger causality strength with a preset strength threshold obtained by training based on system health history data;

[0032] Simultaneously monitor the changing trends of asymmetry indicators over multiple periods;

[0033] When the Granger causality strength is less than the preset strength threshold and the asymmetry index continues to rise monotonically for multiple consecutive analysis cycles, it is determined that the air conditioning system has a hidden performance degradation caused by filter blockage or scale buildup on the heat exchange surface.

[0034] Furthermore, based on the results of the implicit performance degradation assessment, equipment maintenance instructions are generated and control strategies are adjusted, including:

[0035] Based on the determination of the hidden performance degradation caused by filter clogging or scale buildup on the heat exchange surface, equipment maintenance instructions are generated, including filter cleaning reminders and heat exchanger inspection requirements.

[0036] At the same time, the frequency control commands of the central controller to the compressor are adjusted to reduce the upper limit of the compressor's operating frequency and increase the fan speed in order to maintain the temperature stability of the area served by the air conditioning system.

[0037] The beneficial effects of this invention are:

[0038] 1. Through multi-dimensional data fusion analysis and collaborative diagnosis mechanism, the problem of the difficulty in timely detection of hidden performance degradation in air conditioning systems under flat architecture is effectively solved. By collecting operating parameters and environmental parameters in real time, calculating energy efficiency deviation and analyzing the collaborative operation status of equipment, performance degradation characteristics caused by filter blockage or heat exchange surface scaling can be identified at an early stage. The dual verification mechanism of Granger causality analysis and power spectrum asymmetry index is adopted, which significantly improves the accuracy and reliability of fault diagnosis and avoids the misjudgment phenomenon that may occur in traditional single parameter monitoring.

[0039] 2. It realizes the transformation from passive response to proactive early warning. By generating maintenance instructions and adaptively adjusting the control strategy at the early stage of performance degradation, it not only ensures the stability of temperature and humidity control, but also avoids the equipment from operating in a low-energy-efficiency state for a long time. This intelligent diagnostic method based on multi-index joint analysis significantly improves the energy-saving effect and operational safety of the system, and provides an effective means of status monitoring and energy efficiency assurance for flat architecture air conditioning systems. Attached Figure Description

[0040] Figure 1This is a schematic diagram of the structure of an air conditioning control system based on a flat design according to the present invention;

[0041] Figure 2 This is a flowchart for determining whether a system has hidden performance degradation, as described in this invention. Detailed Implementation

[0042] 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.

[0043] Example: Figure 1 A schematic diagram of an air conditioning control system based on a flat design is provided according to the present invention. The air conditioning control system based on a flat design includes:

[0044] The parameter acquisition module is used to collect the operating parameters and corresponding environmental parameters of each device in the system in real time.

[0045] The energy efficiency analysis module is used to calculate the current actual energy efficiency value based on operating parameters and environmental parameters, and compare it with the preset benchmark energy efficiency value to generate energy efficiency deviation data;

[0046] The step loss analysis module is used to identify the period during which the system is in a stable temperature state. It analyzes the step loss rate of energy consumption trajectory by extracting the energy consumption data of multiple devices operating collaboratively during the corresponding maintenance period.

[0047] The joint analysis module is used to analyze the Granger causal relationship between the total energy consumption time series and the key control command time series when the energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, and to calculate the asymmetry index of the instantaneous power spectrum of the system.

[0048] The performance determination module is used to determine that if the Granger causality strength is less than a preset strength threshold and the asymmetry index continues to increase synchronously, the system has hidden performance degradation.

[0049] The result execution module is used to generate equipment maintenance instructions and adjust control strategies based on the results of the implicit performance degradation determination.

[0050] The system collects the operating parameters and corresponding environmental parameters of each device in real time, specifically as follows:

[0051] To achieve accurate perception of the air conditioning system's operating status, it is first necessary to collect real-time operating parameters of each key device in the system, as well as environmental parameters of the environment in which they are located. Operating parameters reflect the electrical and operational status of the equipment itself, while environmental parameters characterize the real-time heat load demand of the area served by the air conditioning system. The simultaneous collection of these two types of parameters provides a data foundation for subsequent energy efficiency calculations and fault diagnosis. In specific implementation, the objects of operating parameter collection clearly include three key devices in the air conditioning system: compressors, fans, and water pumps.

[0052] The acquisition of compressor operating parameters is achieved through current transformers and voltage sensors installed in its power distribution circuit. For example, current transformers and voltage sensors with an accuracy class of 0.5 are used, and the instantaneous values ​​of operating current and voltage are measured and recorded synchronously at a sampling frequency of once per second. Based on the measured current and voltage values, the real-time active power is calculated according to the power calculation formula "active power equals current multiplied by voltage multiplied by power factor". The power factor can be directly read from a power factor table or obtained by calculating the ratio of apparent power to active power. Simultaneously, the current operating frequency command value is obtained in real time from the control command data frames sent from the central controller to the compressor inverter through standard industrial communication protocols such as Modbus-RTU. This frequency command value is the compressor's operating frequency data.

[0053] For the acquisition of operating parameters of fans and water pumps, the same acquisition method and hardware configuration as for compressors are adopted. That is, independent current transformers and voltage sensors are installed in their respective power circuits, and their instantaneous operating current and voltage values ​​are synchronously acquired at the same sampling frequency of once per second. Their active power is then calculated using the same power calculation formula. Similarly, the operating frequency data of fans and water pumps is obtained by parsing the commands sent from the controller to its frequency converter. The operating current, active power, and operating frequency data of all equipment are encapsulated into data packets with a unified timestamp. This timestamp is provided by the real-time clock chip built into the data acquisition unit, with an accuracy down to the millisecond level, ensuring strict synchronization and comparability among the parameters of multiple devices.

[0054] Environmental parameter collection is conducted within the service area of ​​the air conditioning system. High-precision integrated temperature and humidity sensors, such as digital temperature and humidity sensors, are deployed at the geometric center of this area, away from heat sources and air supply / return vents. These sensors measure and record real-time temperature and humidity values ​​at a sampling frequency of once per second. Before deployment, the sensors are calibrated using standard metrology equipment to ensure temperature measurement accuracy is better than ±0.5 degrees Celsius and humidity measurement accuracy is better than ±3%RH, guaranteeing data reliability. The temperature and humidity data collected by the sensors are also tagged with a unified timestamp.

[0055] All operational and environmental parameters are acquired and controlled by a data acquisition unit. This unit features multiple analog input ports and digital communication interfaces for connecting to various sensor and controller networks. Its built-in real-time clock provides millisecond-accurate time stamps for each acquired data point. Operational and environmental parameter data are reliably transmitted to subsequent data processing and storage units via wired communication networks such as RS-485 bus or Ethernet, or wireless communication methods such as ZigBee or LoRa. Data transmission protocols can adopt industry-standard protocols such as Modbus-TCP or MQTT to ensure data integrity and resolvability. The entire acquisition process is performed periodically, with each acquisition cycle strictly adhering to the set sampling frequency, thus ensuring the real-time and continuous monitoring of system status and providing a solid data foundation for subsequent analysis. Before data upload, the data acquisition unit can perform preliminary verification of the raw data, such as checking whether the data exceeds reasonable ranges (e.g., current values ​​should not be negative, and temperature values ​​should be within the expected range), and marking abnormal data points. However, the raw data is still uploaded in its entirety for subsequent analysis and judgment.

[0056] The actual energy efficiency value is calculated based on operating parameters and environmental parameters, and compared with the preset benchmark energy efficiency value to generate energy efficiency deviation data. The specific implementation is as follows:

[0057] After obtaining real-time data on the current, power, and operating frequency of the compressor, fan, and water pump, as well as real-time temperature and humidity data, the current actual energy efficiency value of the air conditioning system is calculated. The current actual energy efficiency value is a key indicator characterizing the instantaneous energy conversion efficiency of the air conditioning system, defined as the ratio between the total cooling capacity generated by the system per unit time and the total power consumed. The calculation process first requires determining the total cooling capacity, which can be estimated through the refrigerant cycle characteristics and air-side parameters of the air conditioning system. For example, for an air-cooled air conditioning system, its total cooling capacity can be calculated by measuring the enthalpy difference between the evaporator outlet and return air, as well as the airflow. Specifically, based on real-time temperature and humidity data, combined with a table of humid air properties or calculation formulas, the specific enthalpy values ​​of the return air and supply air are calculated separately. Then, the airflow on the evaporator side is measured using an airflow sensor. For example, a Pitot tube or thermal anemometer is used to measure the flow velocity at the duct cross-section, and the volumetric airflow is calculated based on the duct cross-sectional area. This volumetric airflow is then multiplied by the air density to obtain the mass airflow. Finally, the total cooling capacity equals the mass airflow multiplied by the specific enthalpy difference between the outlet and return air. Another method is to calculate the total cooling capacity based on the refrigerant mass flow rate and the enthalpy difference between the evaporator inlet and outlet. The refrigerant mass flow rate can be estimated using parameters such as the opening of the electronic expansion valve, compressor speed, and condensing pressure. Next, the total power consumption is calculated. Total power consumption is the sum of the active power consumed by all major energy-consuming devices in the system within the same unit of time; that is, it is directly added together from the active power of the compressor, fan, and water pump measured by the current transformer and voltage sensor. The calculated total cooling capacity is then divided by the total power consumption. The resulting ratio is the current actual energy efficiency value, which is usually a dimensionless number or measured in kilowatts per kilowatt. Its magnitude directly reflects the system's efficiency in converting input electrical energy into cooling capacity.

[0058] After obtaining the current actual energy efficiency value, it needs to be compared with the preset benchmark energy efficiency value to assess the degree of deviation. The preset benchmark energy efficiency value is derived from a large amount of energy efficiency data collected during the system's historical healthy operating conditions. Historical healthy operating conditions refer to a period of time during which the air conditioning system has been confirmed to be fault-free and operating stably, such as a period of 2 to 3 weeks of continuous stable operation after maintenance. During this period, the system continuously calculates and records its actual energy efficiency value, forming a time series of energy efficiency values. This time series data is cleaned to remove outliers caused by drastic changes in operating conditions or short-term faults. The benchmark value is then determined through statistical analysis methods, such as taking the arithmetic mean of all energy efficiency values ​​within this time period as the preset benchmark energy efficiency value. This value is stored in the system's non-volatile memory as a reference benchmark for energy efficiency evaluation.

[0059] The process of generating energy efficiency deviation data is essentially the process of quantifying the degree to which the current actual energy efficiency value deviates from the preset benchmark energy efficiency value. This deviation is expressed as a percentage deviation, calculated using the following formula: subtract the preset benchmark energy efficiency value from the current actual energy efficiency value, divide the difference by the preset benchmark energy efficiency value, and finally multiply the quotient by 100% to convert it to a percentage. The calculated percentage deviation value is the energy efficiency deviation data, which is a signed numerical value. A positive deviation indicates that the current energy efficiency is higher than the historical benchmark level, while a negative deviation indicates that the current energy efficiency is lower than the historical benchmark level. The energy efficiency deviation data is output in real time and transmitted to subsequent analysis processes to determine whether the system's energy efficiency status is abnormal. The entire calculation and comparison process is performed cyclically at fixed time intervals, such as every 5 minutes, to ensure continuous monitoring of the system's energy efficiency status. During the calculation process, the system will perform validity checks on the input data. For example, it will check whether the temperature value is within a reasonable range of -20℃ to +60℃, whether the humidity value is within a range of 0% to 100%, and whether the power value is non-negative. Abnormal data points that exceed the reasonable range will be marked and replaced with the previous valid value to ensure the stability and reliability of the calculation.

[0060] During the period when the identification system is in a stable temperature state, the energy consumption trajectory synchronization rate is analyzed by extracting energy consumption data from multiple devices operating collaboratively during the corresponding maintenance period. Specifically, the implementation is as follows:

[0061] Identifying the period during which the system maintains a stable temperature state is a prerequisite for analyzing energy consumption synergy characteristics. This determination relies on real-time temperature data from the area served by the air conditioning system. Once the determination process begins, the system continuously monitors real-time temperature data collected by temperature and humidity sensors deployed within the service area and compares it with the user-set target temperature value. The preset temperature fluctuation range is an acceptable temperature deviation interval centered on the set temperature value. The size of this range is preset based on the control accuracy and comfort requirements of the air conditioning system; for example, it can be set to ±0.5 degrees Celsius of the set temperature value. The preset duration refers to the shortest time the real-time temperature must remain within the aforementioned fluctuation range. This duration ensures the system truly enters and maintains a stable state, avoiding misjudgments due to short-term temperature fluctuations; for example, the preset duration can be set to 15 minutes. When the system detects that the real-time temperature data remains within the preset temperature fluctuation range (e.g., set value ±0.5℃) for more than the preset duration (e.g., 15 minutes), it determines that the system has entered a stable temperature maintenance period from that moment. This maintenance period will continue until the real-time temperature exceeds the fluctuation range again. The system will record the start and end timestamps of this period and maintain data collection and analysis throughout the entire period.

[0062] After entering a stable temperature maintenance period, the system begins extracting energy consumption data from multiple devices operating collaboratively during this period to analyze their synergy. Specifically, focusing on two key collaborative devices—compressors and fans—the system extracts power consumption data for both compressors and fans from the stored real-time operating parameter database, timestamped throughout the entire maintenance period. The extracted data undergoes strict time alignment to ensure that the data sequences of the two devices have identical time resolution and time points. For example, if the data acquisition frequency is once per second, the extracted data maintains the same frequency, and interpolation is performed to address any data loss caused by communication delays or other reasons. Subsequently, the compressor power consumption data is arranged chronologically to form a compressor power time series; similarly, the fan power consumption data is arranged chronologically to form a fan power time series. These two time series are of equal length, and each data point has a precise timestamp, serving as the direct input data source for subsequent calculations of the energy consumption trajectory synchronization rate.

[0063] After obtaining the compressor power time series and the fan power time series, the energy consumption trajectory step-out rate is quantified by calculating the dynamic time warping distance between them. Dynamic time warping is an algorithm used to measure the similarity between two time series, which can effectively handle situations where there is stretching and curvature on the time axis. The calculation process first requires calculating the Euclidean distance between all data point pairs between the two series. For a compressor power time series of length M and a fan power time series of length N, an M×N matrix is ​​constructed. Each element (i,j) in the matrix represents the Euclidean distance between the i-th point of the compressor series and the j-th point of the fan series. The Euclidean distance is calculated by taking the square root of the square of the difference between the corresponding point values. Subsequently, a cumulative distance matrix is ​​constructed using a dynamic programming algorithm. This matrix has the same dimension as the aforementioned distance matrix, and each element D(i,j) represents the minimum cumulative distance among all possible paths from the sequence starting point (1,1) to the current point pair (i,j).

[0064] The core of dynamic programming is the iterative calculation of the cumulative distance, with the recursive relationship defined as D(i,j)=d(i,j)+min(D(i-1,j),D(i,j-1),D(i-1,j-1)), where d(i,j) is the Euclidean distance between points (i,j), and min represents the minimum value. After filling the entire cumulative distance matrix in this iterative manner, the algorithm traces backward from the bottom right element D(M,N) to the top left element D(1,1) to find a path that minimizes the cumulative distance; this path is the optimal curved path. Finally, the value of the bottom right element D(M,N) of the cumulative distance matrix, i.e., the minimum cumulative distance value, is extracted as the dynamic time warping distance. The value of this dynamic time warping distance is the energy consumption trajectory synchronization rate. The larger the value, the worse the synchronization of the power change trajectory of the compressor and the fan in the time dimension, i.e., the higher the degree of synchronization loss; conversely, the lower the value, the better the synchronization. The entire calculation process is implemented through programming, ensuring that the extracted time series data can be processed automatically and a specific step loss rate value is output for subsequent judgment. During the calculation process, the input data is also normalized to eliminate the influence of dimensions, and reasonable boundary conditions are set to ensure the stability of the calculation.

[0065] When the energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, the Granger causality strength between the total system energy consumption time series and the key control command time series is analyzed, and the asymmetry index of the system's instantaneous power spectrum is calculated. Specifically, the implementation is as follows:

[0066] When energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, the system initiates a deep analysis process to analyze the Granger causality strength between the system's total energy consumption time series and the key control command time series, and calculates the asymmetry index of the system's instantaneous power spectrum. This process first requires confirming the triggering condition: energy efficiency deviation data continuously exceeds the preset deviation threshold for multiple consecutive sampling periods, while the energy consumption trajectory out-of-step rate exhibits a monotonically increasing trend. The preset deviation threshold is determined based on the statistical distribution of historical energy efficiency deviation data under healthy operating conditions. For example, the average historical energy efficiency deviation data plus twice the standard deviation can be used as the threshold, where both the average and standard deviation are calculated based on data from at least 30 days of normal system operation. The determination of a monotonically increasing trend is achieved by analyzing the trend of the energy consumption trajectory out-of-step rate values ​​calculated over the most recent sampling periods. For example, a linear regression method is used to calculate the slope; when the slope is greater than zero and the statistical significance level is less than 0.05, it is considered monotonically increasing. When both conditions are met, the system determines that in-depth analysis is required, and the analysis time is limited to the period during which the temperature stability state has been identified.

[0067] After triggering the analysis, the system constructs the required time series based on the data within the maintenance period. First, it constructs a system total energy consumption time series based on the total system power consumption data within the maintenance period. This total power consumption data is a sequence of the sum of the active power of all power-consuming equipment, such as compressors, fans, and pumps, recorded chronologically within the maintenance period. The active power of each device is calculated by measuring its operating current and operating voltage. Each data point in this time series corresponds to the system total power value at a sampling time, with the time resolution consistent with the original resolution at the time of acquisition, for example, one data point per second. Simultaneously, a key control instruction time series is constructed based on the frequency control commands issued by the central controller to the compressor within the same maintenance period. This series records the frequency setpoint commands sent by the controller to the compressor inverter, also arranged chronologically and strictly synchronized with the system total energy consumption time series. Time synchronization is achieved through a unified timestamp, with timestamp accuracy down to the millisecond level.

[0068] After obtaining the system total energy consumption time series and the key control command time series, a Granger causality test is performed on both to quantify the strength of their causal relationship. The Granger causality test process includes: first, performing a stationarity test on both time series, for example, using the unit root test method; if the series is not stationary, differencing is performed until stationarity is achieved. Then, the optimal lag order p is determined, for example, by calculating the AIC value of the model under different lag orders, and selecting the order that minimizes the AIC value as the optimal lag order; the search range for the lag order is usually set to 1 to 12. Next, two vector autoregressive models are established. The first is a full model, which uses the past p values ​​of the key control command time series and the past p values ​​of the system total energy consumption time series to predict the current value of the system total energy consumption time series; the second is a restricted model, which only uses the past p values ​​of the system total energy consumption time series to predict its current value. The model parameters are estimated using the least squares method. Then, the sum of squared prediction errors of the two models is calculated; the sum of squared prediction errors of the full model is denoted as `full`, and the sum of squared prediction errors of the restricted model is denoted as `restricted`. Finally, the F-statistic is calculated using the formula F=[(restricted-full) / p] / [full / (n-2p-1)], where n is the length of the time series. This F-statistic represents the Granger causality strength; a larger value indicates a stronger predictive ability of the key control commands on the system's total energy consumption.

[0069] Simultaneously, the system performs frequency domain analysis on the total system power consumption data during the maintenance period to calculate the asymmetry index. First, the total system power consumption data is preprocessed, including removing the DC component and applying a window function. A Hanning window with a length of 256 sampling points can be used. Then, a Fast Fourier Transform (FFT) is performed on the preprocessed data to obtain the power spectrum. The FFT points are set to 1024, and the sampling frequency is determined based on the actual data acquisition frequency; for example, if the acquisition frequency is 1Hz, then the sampling frequency is 1Hz. The resulting power spectrum describes the distribution of signal power at different frequencies. Next, the center frequency of the power spectrum is calculated. The formula for the center frequency is the sum of the power values ​​at each frequency point multiplied by their corresponding power values, then divided by the total power value. Then, the third central moment of the power spectrum distribution is calculated. The formula for the third central moment is the cube of the difference between each frequency point and the center frequency multiplied by the power value at that frequency point, then summed over all frequency points and divided by the total power value. Simultaneously, the standard deviation of the power spectrum distribution is calculated. The formula for the standard deviation is: the square of the difference between each frequency point and the center frequency, multiplied by the corresponding power value, summed, divided by the total power value, and then the square root is taken. Finally, the third-order central moment is divided by the cube of the standard deviation to obtain the skewness value, which is the asymmetry index of the instantaneous power spectrum of the system.

[0070] Through the above analysis, the system obtains two key indicators: the Granger causality strength characterizes the degree of influence of control commands on energy consumption, while the asymmetry indicator reflects the distribution characteristics of system power consumption in the frequency domain. During the calculation process, the system performs quality checks on the input data, such as checking for missing values ​​in the time series and supplementing them using linear interpolation; it also checks whether the power spectrum meets the analysis requirements, such as whether the total power value is greater than the preset minimum power threshold, ensuring the reliability of the analysis. The entire analysis process is based on rigorous statistical and signal processing theories, and all calculation parameters are optimized and selected according to the actual data characteristics, ensuring the scientific validity and reliability of the results and providing a quantitative basis for subsequent performance degradation determination.

[0071] Figure 2 The present invention provides a flowchart for determining whether a system exhibits implicit performance degradation. If the Granger causality strength is less than a preset strength threshold and the asymmetry index increases synchronously and continuously, then the system is determined to have implicit performance degradation. The specific implementation is as follows:

[0072] After obtaining the analysis results of Granger causality strength and asymmetry index, the system enters the stage of determining implicit performance degradation. This determination process is based on the joint analysis of the two key indicators. First, the calculated Granger causality strength is compared with a preset strength threshold, which is determined through statistical learning methods based on a large amount of historical data from the system under healthy operating conditions. Specifically, the process of determining the preset strength threshold includes collecting all Granger causality strength values ​​calculated during a historical period when the system is confirmed to be fault-free and operating stably (e.g., 30 consecutive days of stable operation immediately after maintenance), forming a historical data set. Statistical analysis is then performed on this data set to calculate its probability distribution. For example, the 5th percentile of this distribution is taken as the preset strength threshold. This means that when the currently calculated Granger causality strength is lower than this threshold, it indicates that the explanatory power of control commands for energy consumption is significantly lower than the level when the system is healthy. This threshold is stored in the system's non-volatile memory and can be periodically updated based on the accumulation of system operating time, for example, the preset strength threshold is recalculated every quarter to reflect the characteristics of the system changing over time.

[0073] Simultaneously, the system continuously monitors the changing trends of asymmetry indicators over multiple periods. Multiple periods refer to several consecutive analysis cycles, each corresponding to a complete period of stable temperature conditions and subsequent data processing. For example, the system might complete the entire process from data acquisition to indicator calculation every 4 hours. Trend monitoring involves recording the asymmetry indicator values ​​calculated at the end of each analysis cycle to form a time series, which is then subjected to trend analysis. Trend analysis can employ statistical methods, such as calculating the linear regression slope within a certain window period. The window size can be set according to system characteristics, for example, setting the window period to 5 analysis cycles. When the regression slopes calculated over multiple consecutive analysis cycles are all positive and statistically significant, the asymmetry indicator is determined to maintain a monotonically increasing trend. The system records the values ​​of the asymmetry indicators and their changing trends, providing data support for subsequent judgments.

[0074] When both conditions are met simultaneously—that the Granger causality strength is less than a preset threshold and the asymmetry index shows a monotonically increasing trend over multiple consecutive analysis cycles—the system determines that the air conditioning system has a hidden performance degradation. The number of consecutive analysis cycles in the determination process can be configured according to system reliability requirements, for example, it can be set to three consecutive analysis cycles. This specific combination of indicators characterizes a specific type of system performance degradation. A weakening Granger causality strength indicates a decrease in the guiding effect of control commands on energy consumption, while a continuous increase in the asymmetry index reflects a shift in the spectral distribution of system power consumption. The combination of these two factors indicates that there is indeed a performance problem caused by increased fluid resistance or decreased heat exchange efficiency. Based on engineering experience and historical fault data analysis, this specific indicator change pattern is highly correlated with performance degradation caused by filter blockage or fouling on heat exchange surfaces. Therefore, the system identifies this type of problem as the cause of the degradation. The determination result will include information such as the degradation type, severity assessment, and occurrence time. This information is encapsulated into structured determination result data and transmitted to the subsequent result execution module for generating maintenance commands and adjusting control strategies. The entire judgment process employs a fault-tolerant mechanism. For example, when data quality is poor or calculation anomalies occur, the system will mark the results of that analysis period as invalid and exclude them from trend judgment, ensuring the reliability of the judgment results. The system will also verify the judgment results. For instance, when a latent performance degradation is judged, it will check whether there have been any recent maintenance records to avoid misjudgment. At the same time, it will record the system operating parameters at the time of the judgment to provide a reference for subsequent maintenance.

[0075] Based on the results of the implicit performance degradation assessment, equipment maintenance instructions are generated and the control strategy is adjusted. The specific implementation is as follows:

[0076] Once the system determines that there is hidden performance degradation due to filter clogging or scaling on the heat exchange surface, it immediately initiates the corresponding equipment maintenance instruction generation and control strategy adjustment process. The generation of equipment maintenance instructions is based on the specific performance degradation determination result. The system first parses the determination result data packet to obtain the degradation type, severity, and time information, and then generates targeted maintenance instructions based on a preset maintenance knowledge base. The maintenance knowledge base stores maintenance measures corresponding to different fault types. For example, for filter clogging, maintenance measures include checking the filter pressure differential and cleaning or replacing the filter; for scaling on the heat exchange surface, maintenance measures include checking the heat exchanger temperature difference and chemical cleaning or mechanical descaling. The system matches the corresponding maintenance items from the knowledge base according to the degradation type in the determination result and generates a structured equipment maintenance instruction. This instruction includes information such as maintenance items, maintenance requirements, and suggested completion deadlines. For example, it can generate a specific instruction such as "Immediately check the filter cleanliness; if the pressure differential exceeds 50 Pa, it needs to be cleaned or replaced." The urgency of maintenance instructions is determined based on the severity of performance degradation. For example, when the Granger causality strength is more than 30% below a preset threshold, the maintenance instruction is marked as urgent and requires processing within 24 hours. The generated equipment maintenance instructions are displayed to maintenance personnel through a human-machine interface, simultaneously transmitted to a remote monitoring platform via the communication network, and recorded in the maintenance log database.

[0077] While generating equipment maintenance instructions, the system adjusts the control strategy of the centralized controller to adapt to the performance degradation state. The control strategy adjustment mainly targets two key parameters: compressor operating frequency and fan operating speed. For the compressor operating frequency, the system calculates a new upper limit value based on the degree of performance degradation. For example, it multiplies the original upper limit value by a degradation coefficient, which is determined based on the proportion of Granger causality strength below a preset threshold. The specific calculation method is: degradation coefficient = 1 - (preset threshold - current threshold) / preset threshold × adjustment factor, where the adjustment factor ranges from 0.5 to 1.0, determined based on system type and experience. The new upper limit value of the operating frequency is sent to the compressor inverter via the control bus to ensure that the compressor does not operate at excessively high frequencies, thus exacerbating performance degradation. For the fan operating speed, the system calculates the target speed increase value based on the need to maintain temperature stability. The magnitude of the increase value is positively correlated with the degree of performance degradation. For example, for every 10% drop in Granger causality strength below the threshold detected, the fan speed baseline value is increased by 3% to 5% of the rated speed. The speed adjustment command is sent to the fan inverter via analog output or communication protocol. These adjustments ensure that the air conditioning system can maintain a stable temperature in the service area even when performance is degraded, while avoiding excessive wear and tear on the equipment.

[0078] All control strategy adjustment parameters are stored in non-volatile memory and are subject to safety boundary limits. For example, the lower limit of compressor frequency is no less than 30% of the rated frequency, and the upper limit of fan speed is no more than 115% of the rated speed, ensuring that the equipment operates within a safe range. The system continuously monitors the operational effects after adjustments, evaluating the effectiveness of the control strategy adjustment by comparing indicators such as temperature fluctuation range and energy consumption changes before and after the adjustment. If temperature fluctuations still exceed the allowable range, the system will further optimize the control parameters, such as continuing to increase the fan speed or decrease the upper limit of compressor frequency in 5% increments, until the system reaches a stable state again. All control strategy adjustment operations are recorded in the operation log, including adjustment time, adjustment parameters, and values ​​before and after the adjustment, providing data support for subsequent maintenance and system optimization. The system also establishes an adjustment effect evaluation mechanism. For example, within 24 hours after the control strategy adjustment, temperature stability data is collected every 2 hours, the standard deviation of temperature fluctuation is calculated, the actual effect of the control strategy adjustment is evaluated, and a decision is made based on the evaluation results as to whether further adjustments to the control parameters are needed.

[0079] In the embodiments, unless otherwise specified, "system" is used to refer to an air conditioning system.

[0080] This embodiment adopts a flat architecture that directly connects and controls all field devices through a centralized controller, eliminating the intermediate control layer in traditional distributed control and realizing centralized management of device control. The centralized controller establishes communication connections directly with actuators such as compressors, fans, and pumps, as well as sensor devices, through a unified control bus, realizing direct interaction between data acquisition and device control, reducing information transmission links, improving system response speed and control efficiency, and providing a foundation for system-level energy efficiency optimization and fault diagnosis through global data sharing.

[0081] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0082] 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.

[0083] 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.

[0084] Those skilled in the art will 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.

[0085] In the several 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.

[0086] 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.

[0087] In addition, the functional modules in the various 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.

[0088] 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.

[0089] 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.

[0090] 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. An air conditioning control system based on a flattened design, characterized by, include: The parameter acquisition module is used to collect the operating parameters and corresponding environmental parameters of each device in the system in real time. The energy efficiency analysis module is used to calculate the current actual energy efficiency value based on operating parameters and environmental parameters, and compare it with the preset benchmark energy efficiency value to generate energy efficiency deviation data; The step loss analysis module is used to identify the period during which the system is in a stable temperature state. It analyzes the step loss rate of energy consumption trajectory by extracting the energy consumption data of multiple devices operating collaboratively during the corresponding maintenance period. The joint analysis module is used to analyze the Granger causal relationship between the total energy consumption time series and the key control command time series when the energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, and to calculate the asymmetry index of the instantaneous power spectrum of the system. The performance determination module is used to determine that if the Granger causality strength is less than a preset strength threshold and the asymmetry index continues to increase synchronously, the system has hidden performance degradation. The result execution module is used to generate equipment maintenance instructions and adjust control strategies based on the results of the implicit performance degradation determination.

2. The air conditioner control system based on the flattened design according to claim 1, wherein, Real-time acquisition of operating parameters and corresponding environmental parameters of each device in the system, including: The current, power, and operating frequency data of the compressor, fan, and water pump in the air conditioning system are periodically and synchronously collected as operating parameters. Real-time temperature and humidity data of the air conditioning system service area are collected synchronously as environmental parameters.

3. The flat design based air conditioner control system according to claim 2, characterized in that, The current actual energy efficiency value is calculated based on operating parameters and environmental parameters, and compared with the preset benchmark energy efficiency value to generate energy efficiency deviation data, including: Based on real-time collected operating and environmental parameters, the ratio of the total cooling capacity to the total power consumption of the air conditioning system per unit time is calculated as the current actual energy efficiency value. The current actual energy efficiency value is compared with the energy efficiency benchmark value obtained from the system's historical healthy operating conditions, and the percentage deviation between the actual energy efficiency value and the preset benchmark energy efficiency value is calculated to generate energy efficiency deviation data.

4. The flat design based air conditioner control system according to claim 3, wherein During the period when the identification system is in a stable temperature state, the energy consumption trajectory synchronization rate is analyzed by extracting energy consumption data from multiple devices operating collaboratively during the corresponding maintenance period, including: When the real-time temperature of the area served by the air conditioning system remains within the preset temperature fluctuation range and exceeds the preset duration, the system is determined to enter a period of temperature stabilization. Extract the power consumption data of the compressor and fan that operate in coordination during the maintenance period, and generate the compressor power time series and fan power time series respectively; Calculate the dynamic time warping distance between the compressor power time series and the fan power time series, and use the dynamic time warping distance as the energy consumption trajectory step loss rate.

5. The flat design based air conditioner control system according to claim 4, wherein, Calculating the dynamic time warping distance between the compressor power time series and the fan power time series includes: using a dynamic programming algorithm to calculate the Euclidean distance between all data point pairs between the compressor power time series and the fan power time series and constructing a cumulative distance matrix; finding the optimal path from the top left corner to the bottom right corner in the cumulative distance matrix and using the minimum cumulative distance value on the optimal path as the energy consumption trajectory step loss rate.

6. The flat design based air conditioner control system according to claim 4, wherein When energy efficiency deviation data continuously exceeds the preset deviation threshold and the energy consumption trajectory out-of-step rate continues to increase, analyze the Granger causality strength between the system's total energy consumption time series and the key control command time series, and calculate the asymmetry index of the system's instantaneous power spectrum, including: When the energy efficiency deviation data exceeds the preset deviation threshold for multiple consecutive sampling periods and the energy consumption trajectory out-of-step rate shows a monotonically increasing trend, a system total energy consumption time series is constructed based on the system total power consumption data during the maintenance period, and a key control command time series is constructed based on the frequency control commands issued by the central controller to the compressor during the maintenance period. A Granger causality test is performed on the system total energy consumption time series and the key control command time series, and the F statistic value in the test result is used as the strength of the Granger causality relationship. The power spectrum is obtained by performing a fast Fourier transform on the total power consumption data of the system during the maintenance period. The skewness of the corresponding power spectrum distribution relative to its center frequency is calculated as an index of the asymmetry of the instantaneous power spectrum of the system.

7. The flat design based air conditioner control system according to claim 6, wherein The Granger causality test for the total system energy consumption time series and the critical control command time series includes: predicting the current value of the total system energy consumption time series using past values ​​of the critical control command time series, and predicting the current value of the total system energy consumption time series without using past values ​​of the critical control command time series; comparing the error of the prediction results of the total system energy consumption time series under the two prediction methods; and calculating the F statistic as the strength of Granger causality based on the difference in the error.

8. The flat design based air conditioner control system according to claim 6, wherein, The calculation of the skewness of the power spectrum distribution corresponding to the total power consumption data of the system during the maintenance period relative to its center frequency includes: performing a fast Fourier transform on the total power consumption data of the system to obtain the power spectrum, calculating the center frequency of the power spectrum, and then calculating the ratio of the third central moment of the power spectrum distribution to the cube of the standard deviation, using the ratio as an indicator of the asymmetry of the instantaneous power spectrum of the system.

9. The flat design based air conditioner control system according to claim 6, wherein, If the Granger causality strength is less than a preset strength threshold and the asymmetry index continues to increase synchronously, then the system is determined to have hidden performance degradation, including: Compare the Granger causality strength with a preset strength threshold obtained by training based on system health history data; Simultaneously monitor the changing trends of asymmetry indicators over multiple periods; When the Granger causality strength is less than the preset strength threshold and the asymmetry index continues to rise monotonically for multiple consecutive analysis cycles, it is determined that the air conditioning system has a hidden performance degradation caused by filter blockage or scale buildup on the heat exchange surface.

10. The flat design based air conditioner control system according to claim 9, wherein, Based on the results of the implicit performance degradation assessment, equipment maintenance instructions are generated and control strategies are adjusted, including: Based on the determination of the hidden performance degradation caused by filter clogging or scale buildup on the heat exchange surface, equipment maintenance instructions are generated, including filter cleaning reminders and heat exchanger inspection requirements. At the same time, the frequency control commands of the central controller to the compressor are adjusted to reduce the upper limit of the compressor's operating frequency and increase the fan speed in order to maintain the temperature stability of the area served by the air conditioning system.

Images