A fan coil unit temperature control device and temperature control method

Abstract

This invention discloses a fan coil unit temperature control device and method, relating to the field of temperature control technology. The invention connects a microcontroller unit (MCU) and an NB-IoT chip to a remote MQTT cloud server. The NB-IoT chip uploads the current room temperature, outlet air temperature, weather conditions, and room load model to the remote MQTT cloud server. Based on the current parameters and the room load model, the remote MQTT cloud server predicts the room temperature and humidity at the next moment and controls the fan coil unit via the MCU. In other words, this invention connects the MCU to the cloud server via the MQTT protocol, allowing for precise temperature control of the fan coil unit in complex buildings.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology, and in particular to a fan coil unit temperature control device and temperature control method. Background Technology

[0002] In large public buildings, the air conditioning terminal fan coil system is an essential device to ensure the normal production and life of users. Usually, the operation of the fan coil unit is controlled by the fan coil thermostat in the room. The thermostat senses the room temperature through its own temperature sensor and outputs control power to control the fan speed and the opening and closing of the coil water valve.

[0003] Currently, the control logic of fan coil unit thermostats on the market is that when the indoor temperature exceeds the set temperature threshold, the coil water valve opens, and the control system outputs the corresponding power according to the magnitude of the exceedance to control the high-speed, medium-speed, and low-speed operation of the fan.

[0004] However, current fan coil unit thermostats can only control the fan at simple high, medium and low speeds, making it difficult to provide precise control over the fan coil unit to cope with complex building temperature control. Summary of the Invention

[0005] This invention provides a fan coil unit temperature control device and method, which can solve the problem in the prior art that it is difficult to perform precise control of fan coil units to cope with complex building temperature control.

[0006] This invention provides a fan coil unit temperature control device, including a microcontroller unit (MCU), a Flash chip, an NB-IoT chip, and a remote MQTT cloud server; The Flash chip is connected to the microcontroller unit (MCU) and is used to store the collected current room temperature, air outlet temperature, and weather conditions. The microcontroller unit (MCU) is used to construct the room load model and the logic control of the fan coil unit; The NB-IoT chip is connected to the microcontroller unit (MCU) and the remote MQTT cloud server. The NB-IoT chip is used to upload the current room temperature, air outlet temperature, weather conditions, and room load model to the remote MQTT cloud server. The remote MQTT cloud server is used to predict the room temperature and humidity at the next moment based on the current room temperature, air outlet temperature, weather conditions, and room load model, and inputs the predicted temperature and humidity into the microcontroller unit (MCU). The MCU controls the fan coil unit according to the predicted temperature and humidity.

[0007] Preferably, it also includes a temperature and humidity sensor, an outlet air temperature sensor, a system power supply, a button circuit, an output circuit, and a display screen; The temperature and humidity sensor is connected to the Flash chip. The temperature and humidity sensor is used to collect the current room temperature and humidity and store them in the Flash chip. The outlet air temperature sensor is connected to the Flash chip. The outlet air temperature sensor is used to collect the outlet air temperature of the fan coil unit and store it in the Flash chip. The button circuit is connected to the microcontroller unit (MCU). The button circuit changes the temperature and humidity set by the MCU through a combination of a touch chip and a touch spring. The output circuit is connected to the microcontroller unit (MCU), and the output circuit is used to control the switching of the fan and the water valve of the coil according to the temperature and humidity set by the microcontroller unit (MCU). The display screen is connected to the microcontroller unit (MCU), and the display screen is used to display the current temperature and humidity, set temperature, wind speed, mode, on / off status, and NB-IoT signal status. The system power supply is used to convert external AC power into DC power to power the control device.

[0008] Preferably, the room load model is as follows: ; in: This indicates the room temperature to be calculated at the next moment. Indicates the building envelope coefficient; Indicates the room's base temperature; Indicates the thermal characteristic coefficient of a room; Indicates the cooling capacity of the air conditioner; Represents the time-varying matrix of the room; This represents the room's energy consumption variable.

[0009] Air conditioning cooling capacity This is obtained through the coil's air volume and air temperature, and is expressed as: ; in: This represents the specific heat capacity at constant pressure of air. Indicates the density of air; Indicates the air volume supplied by the air conditioner; This indicates the air conditioner's air supply temperature.

[0010] and A parameter indicating the effect of heat generated by people and indoor electrical equipment on room temperature; Time-varying matrix The coefficient for the variation of indoor energy consumption within a day is expressed as: ; The time-varying matrix is ​​obtained using indoor energy consumption metering values. This represents basic indoor energy consumption.

[0011] Preferably, the parameters , and The acquisition includes: Using the XGBoost model to adjust parameters , and The XGBoost model is represented as follows: ; The room temperature data, supply air temperature data, and supply air volume data were cleaned, grouped, and coded to form a training set. The objective function was then established as follows: ; The parameters of XGBoost are continuously trained using the training set, and the parameters are calculated using the trained XGBoost model. , and value.

[0012] Preferably, the microcontroller unit (MCU) performs logic control of the fan coil unit, including: The microcontroller unit (MCU) reads the current room temperature, air outlet temperature, and weather conditions stored in the Flash chip. Based on the current room temperature, air outlet temperature, weather conditions, and the set fuzzy control algorithm, the control quantities for fan speed and valve opening are calculated. The control input is converted into a PWM signal or a relay signal to control the operation of the fan and valve.

[0013] This invention also provides a fan coil unit temperature control method, applied to a fan coil unit temperature control device as described above, comprising the following steps: The microcontroller unit (MCU) extracts the current room temperature, air outlet temperature, and weather status stored in the Flash chip, and uploads the current room temperature, air outlet temperature, weather status, and the current air volume of the fan coil unit to a remote MQTT cloud server. The remote MQTT cloud server inputs the current room temperature, air outlet temperature, weather conditions, and the current air volume of the fan coil unit into the room load model to calculate the indoor temperature at the preset time point. The remote MQTT cloud server performs a weighted calculation by combining the real-time room temperature with the set room temperature to obtain a weighted temperature; and inputs the weighted temperature and the indoor temperature at a preset time point into the fuzzy control algorithm, and outputs the fan speed change value based on the fuzzy control rules in the fuzzy control algorithm. The fan speed change value is input to the microcontroller unit (MCU), which controls the temperature of the fan coil unit based on the fan speed change value.

[0014] Preferably, the step of diagnosing abnormalities in the fan coil unit temperature control device during fan coil unit temperature control includes the following steps: The Dynamic Time Warping (DTW) detection algorithm is used to diagnose anomalies in fan coil units. The DTW algorithm calculates the similarity between the actual operating data sequence and the model prediction sequence. The actual operating data sequence refers to the actual operating data of the fan coil unit within a preset time period, and the model prediction sequence refers to the operating data of the fan coil unit within the preset time period predicted by the fuzzy control algorithm. The similarity is expressed as: ; Where: the actual operating data sequence is represented as The model's predicted sequence is represented as ; λ ( i , j The similarity distance between two time series data is represented as: ; When the similarity D is less than the preset threshold, the fan coil unit is judged to be abnormal.

[0015] The fan coil unit temperature control device and method provided in this invention have the following advantages compared with the prior art: This invention connects a microcontroller unit (MCU) and an NB-IoT chip to a remote MQTT cloud server. The NB-IoT chip uploads the current room temperature, air outlet temperature, weather conditions, and room load model to the remote MQTT cloud server. Based on the current parameters and the room load model, the remote MQTT cloud server predicts the room temperature and humidity at the next moment and controls the fan coil units through the MCU. In other words, this invention connects the MCU to the cloud server via the MQTT protocol, and uses the cloud server to finely control the fan coil units, thereby enabling precise temperature control in complex buildings.

[0016] Moreover, this invention can detect abnormal operating conditions such as damaged or stuck water valves and blocked fans in fan coil units, thereby improving the efficiency of fan coil unit utilization. Attached Figure Description

[0017] Figure 1 A schematic diagram of a conventional fan coil unit temperature controller provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the temperature control architecture of a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 3 A schematic diagram of a circuit principle for a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 4 A schematic diagram of a type II circuit for a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 5 A schematic diagram of three types of circuit principles for a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 6 A schematic diagram of a PCB design for a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of a type II PCB design for a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of a fan coil unit temperature control device provided in an embodiment of the present invention; Figure 9 This is a schematic diagram of the control flow of a fan coil unit temperature control device and temperature control method provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of the temperature control process of a fan coil unit temperature control device and temperature control method provided in an embodiment of the present invention; The components include: 1. air duct, 2. fan, 3. coil, 4. water valve, and 5. thermostat. Detailed Implementation

[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] See Figure 1 This is a schematic diagram of a traditional fan coil thermostat. Its control logic is as follows: when the indoor temperature exceeds the set temperature threshold, the thermostat 5 controls the coil 3 and water valve 4 to open. The control system outputs the corresponding power according to the magnitude of the temperature exceeding the threshold to control the fan 2 to run at high speed, medium speed, or low speed.

[0020] However, with the increasing demand from users for refined room environment management and the market's higher requirements for product energy efficiency, traditional fan coil unit controllers can only achieve coarse temperature control, which presents the following problems: ① It is difficult to precisely control the room temperature, which leads to fluctuations in the room temperature.

[0021] ② Room environmental information cannot be centrally managed and controlled, and building managers cannot understand the full picture of temperature distribution inside the building, resulting in poor management efficiency and difficulty in taking further energy-saving measures.

[0022] ③ Many rooms in the building can only be controlled independently and cannot be linked with the cold source. The cold source lacks an actual model of the terminal cooling load, resulting in an imbalance between the supply and demand of cooling.

[0023] To address the problems found in traditional fan coil unit controllers, this invention proposes a fan coil unit temperature control device (hereinafter referred to as a temperature controller), such as... Figure 2 As shown, the thermostat includes a microcontroller unit (MCU) [1] for logic control of the fan coil system and modeling calculation of room load; a flash chip [6] for storing collected room temperature, air outlet temperature, and weather status information so that the MCU can retrieve the data for modeling calculation; an NB-IoT chip [2] for connecting to the NB network and connecting to a remote MQTT cloud server to upload the current room temperature and humidity, room model parameters, subscribe to outdoor weather information and send it to the MCU for processing, and use a temperature and humidity sensor to measure indoor temperature and humidity; and a VA code display [4] for displaying... Current temperature and humidity, set temperature, wind speed, mode, switch and NB-IOT signal status; button circuit [5], through the combination of touch chip and touch spring, allows users to change the status and design value of the control system, the MCU drives the output circuit [7] according to the control logic, so that the output relays are combined in different ways, thereby outputting power to the fan and coil water valve; air outlet temperature sensor [9], used to measure the air outlet temperature of the coil; system power supply [8], used to convert the external 220V AC power into 5V500mA output power through an isolated AC / DC converter.

[0024] like Figure 3 , Figure 4 and Figure 5 The diagram shown illustrates several circuit designs of this invention; as follows: Figure 6 and Figure 7 The diagram shown illustrates the soldering of the circuit design onto a PCB board. Figure 8 The image shown is a schematic diagram of the product formed after welding according to the present invention.

[0025] In a specific experiment, the present invention uploads the equipment operating parameters and room environment parameters to the backend server via the MQTT protocol. The uploaded parameters are shown in Table 1.

[0026] Table 1 Operating parameters and room environment parameters After the temperature controller's MCU collects and processes the data, it converts it into JSON format and publishes it to the background MQTT server using a fixed location name as the topic in the MQTT protocol. At the same time, the temperature controller also subscribes to the bound topic from the MQTT server backend, receives subscription messages, parses them according to the parameter reference table shown in Table 2 through the MCU program, and executes the instructions. The subscription message format is a JSON string.

[0027] Table 2 Parameter Comparison Table The thermostat program is divided into 6 tasks, managed by the FreeRTOS system. The program tasks are: temperature and humidity acquisition program, temperature control program, display program, interactive program, communication program, and load modeling program.

[0028] 1. Temperature and humidity acquisition program.

[0029] like Figure 9 The diagram shows a flowchart of temperature and humidity acquisition via IIC communication between the MCU and the onboard SHT30 sensor to read the temperature and humidity information collected by the sensor.

[0030] The program first initializes the I2C peripheral working status of the MCU, then the MCU sends a data read request to the SHT30. After receiving the acknowledgment bit, the SHT30 sends the status words of temperature and humidity values ​​to the MCU as SRH and ST, respectively, which are 16-bit unsigned integers. Then the MCU calculates the actual temperature and humidity values ​​using the following conversion formula.

[0031] .

[0032] 2. Temperature control program The temperature control program calculates the results using a room load model, continuously optimizes the output, and finally outputs a control signal to control the temperature. The process is as follows: Figure 10 As shown.

[0033] First, the MCU inputs the indoor heat load parameters along with the current wind speed, outlet air temperature, and indoor temperature and humidity parameters into the room load model to predict the indoor temperature 5 minutes later. The real-time temperature and the indoor temperature setpoint are weighted and calculated together with the predicted temperature and input into the fuzzy control algorithm. According to the fuzzy control rules, the output wind speed change value is used to control the fan coil unit. Finally, the room temperature quickly reaches the set value and enters a steady state, greatly reducing room temperature fluctuations and thus improving human comfort.

[0034] 3. Display and interactive program.

[0035] The user operates the thermostat via buttons and displays the thermostat information on the screen. The operation logic is as follows: After pressing the power button, the screen will display the current indoor temperature and humidity, thermostat operating mode, fan speed setting, and NB signal status. The operating modes "ventilation", "cooling", and "heating" can be set by pressing the mode button. The fan speed settings "automatic", "low", "medium", and "high" can be set by pressing the fan speed button. The set temperature can be set using the up and down buttons. After pressing the up and down adjustment buttons, the screen will display the set temperature value.

[0036] 4. Communication program.

[0037] The communication program packages the thermostat data according to the communication data structure, publishes topics via the MQTT protocol, and starts publishing instructions in a cyclic interrupt manner, executing once every minute to send the data to the cloud platform for parsing. The instructions given by the cloud platform will also be given in the form of published topics. The thermostat subscribes to the topic, receives the instructions, parses them, and executes the commands.

[0038] 5. Load modeling program.

[0039] The thermostat is set with default parameter values ​​at the factory. When using it for the first time, the parameter values ​​should be set according to the actual room information. The parameter values ​​are shown in Table 3.

[0040] Table 3 Parameter Values ​​Set The thermostat records its normal operating data in the flash chip. The control program calibrates the room load model once a week. The calibration uses a room load modeling algorithm, which sends room information parameters and thermostat operating data into the model for training. Finally, the calibrated room load model parameters are uploaded to the cloud server.

[0041] Specifically, the modeling of the room load model includes: Room load models play a crucial role in the accuracy and stability of room temperature control, and can be used to overcome the hysteresis of temperature control. Room temperature variations are related to the room envelope, materials, air conditioning cooling capacity, and factors such as occupants and electrical appliances. A room load model can be expressed as:

[0042] .

[0043] in: This indicates the room temperature to be calculated at the next moment. Indicates the building envelope coefficient; Indicates the room's base temperature; Indicates the thermal characteristic coefficient of a room; Indicates the cooling capacity of the air conditioner; Represents the time-varying matrix of the room; This represents the room's energy consumption variable.

[0044] Air conditioning cooling capacity This can be obtained from the coil air volume and supply air temperature, expressed as: .

[0045] in: This represents the specific heat capacity at constant pressure of air. Indicates the density of air; Indicates the air volume supplied by the air conditioner; This indicates the air conditioner's air supply temperature.

[0046] and The influence of human activity and indoor electrical equipment heating on room temperature is characterized by a pattern that is highly correlated with time in public buildings; therefore, a time-varying matrix is ​​used. The coefficient for the variation of indoor energy consumption within a day can be expressed as: .

[0047] This time-varying matrix is ​​generated using indoor energy consumption metering values. The indoor basic energy consumption is estimated by the indoor power distribution capacity.

[0048] In the room load model These are unknown parameters that require estimation using a pattern recognition algorithm. This invention uses the XGBoost algorithm for parameter identification. XGBoost is an algorithm based on gradient boosting trees, an ensemble learning method that iteratively trains a series of decision trees, attempting to correct the error of the previous iteration in each iteration, ultimately combining these weak learners into a strong learner. The XGBoost model formula can be expressed as: .

[0049] After cleaning the original data, it is grouped and coded to form a training set. The objective function is then established. .

[0050] The parameters of XGBoost are continuously trained using the training set, and finally, the XGBoost model can estimate the... The value is then substituted into... The formula forms a room load model; the dynamic parameters in the model are the base temperature, supply air temperature and supply air volume. When the thermostat controls the temperature, the room temperature needs to be estimated. By inputting the dynamic parameters into the model, the predicted room temperature value at the next moment can be obtained, thereby cooperating with the control algorithm to accurately control the room temperature.

[0051] Specifically, the diagnosis of room anomalies includes: When a fan coil unit malfunctions, such as no airflow from the vents or a tripping fault, it is easily detected by the user. However, when the fan coil unit is aging or operating abnormally, such as a damaged or stuck water valve or a blocked fan, the problem may not be enough to trigger a trip or be difficult for the user to detect. In such cases, an anomaly diagnosis algorithm is needed for detection.

[0052] A room load model can serve as baseline data to check actual operating data. If the difference between the two exceeds a set threshold, it can alert users to check. However, detection errors from single-point data may not indicate equipment malfunction, and room temperature changes can lag to varying degrees due to different operating patterns of room occupants and equipment. Therefore, this invention uses the DTW (Time-Digital Wave) detection algorithm for anomaly detection. The DTW algorithm calculates the similarity between two time-series data. If the actual operating data sequence is represented as... The model predicts the sequence as follows: Then the similarity distance can be expressed as: .

[0053] Their similarity can be expressed as: .

[0054] Abnormal data is diagnosed by determining whether the similarity D is below a threshold.

[0055] This invention optimizes the room temperature change curve, enabling the room temperature to quickly reach the set value and enter a steady state with minimal temperature fluctuations, thus improving comfort. Furthermore, by linking the thermostat with a cloud server, this invention facilitates unified management and maintenance. Finally, this invention can detect abnormal operating conditions of fan coil units through an anomaly diagnosis algorithm.

[0056] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A fan coil unit temperature control device, characterized in that, include: Microcontroller unit (MCU), Flash chip, NB-IoT chip, and remote MQTT cloud server; The Flash chip is connected to the microcontroller unit (MCU) and is used to store the collected current room temperature, air outlet temperature, and weather conditions. The microcontroller unit (MCU) is used to construct the room load model and the logic control of the fan coil unit; The NB-IoT chip is connected to the microcontroller unit (MCU) and the remote MQTT cloud server. The NB-IoT chip is used to upload the current room temperature, air outlet temperature, weather conditions, and room load model to the remote MQTT cloud server. The remote MQTT cloud server is used to predict the room temperature and humidity at the next moment based on the current room temperature, air outlet temperature, weather conditions, and room load model, and inputs the predicted temperature and humidity into the microcontroller unit (MCU). The MCU controls the fan coil unit according to the predicted temperature and humidity.

2. The fan coil unit temperature control device according to claim 1, characterized in that, Also includes: Temperature and humidity sensor, outlet air temperature sensor, system power supply, button circuit, output circuit and display screen; The temperature and humidity sensor is connected to the Flash chip. The temperature and humidity sensor is used to collect the current room temperature and humidity and store them in the Flash chip. The outlet air temperature sensor is connected to the Flash chip. The outlet air temperature sensor is used to collect the outlet air temperature of the fan coil unit and store it in the Flash chip. The button circuit is connected to the microcontroller unit (MCU). The button circuit changes the temperature and humidity set by the MCU through a combination of a touch chip and a touch spring. The output circuit is connected to the microcontroller unit (MCU), and the output circuit is used to control the switching of the fan and the water valve of the coil according to the temperature and humidity set by the microcontroller unit (MCU). The display screen is connected to the microcontroller unit (MCU), and the display screen is used to display the current temperature and humidity, set temperature, wind speed, mode, on / off status, and NB-IoT signal status. The system power supply is used to convert external AC power into DC power to power the control device.

3. The fan coil unit temperature control device according to claim 1, characterized in that, The room load model is as follows: ； in: This indicates the room temperature to be calculated at the next moment. Indicates the building envelope coefficient; Indicates the room's base temperature; Indicates the thermal characteristic coefficient of a room; Indicates the cooling capacity of the air conditioner; Represents the time-varying matrix of the room; Represents the room's energy consumption variable; Air conditioning cooling capacity This is obtained through the coil's air volume and air temperature, and is expressed as: ； in: This represents the specific heat capacity at constant pressure of air. Indicates the density of air; Indicates the air volume supplied by the air conditioner; Indicates the air conditioner's air supply temperature; and A parameter indicating the effect of heat generated by people and indoor electrical equipment on room temperature; Time-varying matrix The coefficient for the variation of indoor energy consumption within a day is expressed as: ； The time-varying matrix is ​​obtained using indoor energy consumption metering values. This represents basic indoor energy consumption.

4. The fan coil unit temperature control device according to claim 3, characterized in that, The parameters , and The acquisition includes: Using the XGBoost model to adjust parameters , and The XGBoost model is represented as follows: ； The room temperature data, supply air temperature data, and supply air volume data were cleaned, grouped, and coded to form a training set. The objective function was then established as follows: ； The parameters of XGBoost are continuously trained using the training set, and the parameters are calculated using the trained XGBoost model. , and value.

5. The fan coil unit temperature control device according to claim 1, characterized in that, The microcontroller unit (MCU) performs logic control of the fan coil unit, including: The microcontroller unit (MCU) reads the current room temperature, air outlet temperature, and weather conditions stored in the Flash chip. Based on the current room temperature, air outlet temperature, weather conditions, and the set fuzzy control algorithm, the control quantities for fan speed and valve opening are calculated. The control input is converted into a PWM signal or a relay signal to control the operation of the fan and valve.

6. A fan coil unit temperature control method, applied to the fan coil unit temperature control device according to any one of claims 1 to 5, comprising the following steps: The microcontroller unit (MCU) extracts the current room temperature, air outlet temperature, and weather status stored in the Flash chip, and uploads the current room temperature, air outlet temperature, weather status, and the current air volume of the fan coil unit to a remote MQTT cloud server. The remote MQTT cloud server inputs the current room temperature, air outlet temperature, weather conditions, and the current air volume of the fan coil unit into the room load model to calculate the indoor temperature at the preset time point. The remote MQTT cloud server performs a weighted calculation by combining the real-time room temperature with the set room temperature to obtain a weighted temperature; and inputs the weighted temperature and the indoor temperature at a preset time point into the fuzzy control algorithm, and outputs the fan speed change value based on the fuzzy control rules in the fuzzy control algorithm. The fan speed change value is input to the microcontroller unit (MCU), which controls the temperature of the fan coil unit based on the fan speed change value.

7. The fan coil unit temperature control method according to claim 6, characterized in that, When controlling the temperature of the fan coil unit, the abnormal diagnosis of the fan coil unit temperature control device includes the following steps: The Dynamic Time Warping (DTW) detection algorithm is used to diagnose anomalies in fan coil units. The DTW algorithm calculates the similarity between the actual operating data sequence and the model prediction sequence. The actual operating data sequence refers to the actual operating data of the fan coil unit within a preset time period, and the model prediction sequence refers to the operating data of the fan coil unit within the preset time period predicted by the fuzzy control algorithm. The similarity is expressed as: ； Where: the actual operating data sequence is represented as The model's predicted sequence is represented as ; λ ( i , j The similarity distance between two time series data is represented as: ； When the similarity D is less than the preset threshold, the fan coil unit is judged to be abnormal.

Images