CN101140450B - Energy-saving thermal comfort controller and control method - Google Patents

Abstract

The invention discloses an energy-saving thermal comfort controller and a control method for a heating, ventilating and air conditioning system based on a thermal comfort index, energy-saving control and a neural network algorithm. The controller calculates and displays a thermal comfort index based on the air temperature and humidity measured by the sensor, the wind speed, the radiation temperature, and the user's activity level and wearing condition, which are calculated according to the input power of the blower, and the user's activity level and wearing condition set or selected by the user as default values. The energy-saving control method divides the heating, ventilating and air conditioning system into three operation stages of quick cooling/heating, comfortable operation and energy-saving operation, and provides corresponding comfortable index set values. The neural network algorithm is used for calculating according to the deviation of the set value and the actual value of the comfort index and outputting a control signal, and the heating, ventilation and air conditioning system is controlled through the direct current driving circuit and the alternating current driving circuit respectively, so that the highest comfort degree and the lowest energy consumption are realized.

Description

Energy-saving thermal comfort controller and control method

technical field

The present invention generally relates to a controller and a control method of an indoor climate change device, in particular to a controller and a control method for an HVAC system, so as to provide users with satisfactory energy-saving thermal comfort control effects.

Background technique

At present, HVAC systems have been widely used in daily work and life. The controller of the existing HVAC system usually adopts the constant temperature control method, that is, the user first sets the temperature value and the fan gear, and then adopts the on-off control mode, the proportional-integral control mode or the currently popular fuzzy control mode to control the temperature. The indoor temperature is adjusted, and some advanced or large-scale equipment also adjusts the indoor humidity at the same time. When the indoor air temperature is adjusted to the set value or a certain range, the HVAC system stops cooling/heating, and only the fan continues to work; when the air temperature deviates from the preset temperature value within a certain range, the HVAC system starts cooling again / heating function, so as to keep the indoor air temperature near the temperature set point.

However, with the continuous development of HVAC technology, users have higher and higher requirements for thermal comfort and system energy consumption. The original control system that simply adjusts and maintains the air temperature near the preset temperature value cannot Provides the highest thermal comfort for the user. According to the concept of thermal comfort, the thermal comfort state of the human body is based on the thermodynamic balance between the human body and the surrounding environment. The relationship between thermal comfort, so as to quantitatively express the influence of the compound factors of the warm environment on the human body. At present, the main thermal comfort index includes the predicted warm sensation (Predicted Mean Vote, PMV) index (also called thermal comfort index in this paper) proposed by P.O. Fanger (P.O.Fanger) and has become the ISO standard Satisfaction percentage (Predicted Percentage of Dissatisfied, PPD), and the Effective Temperature (ET) index widely used in the American Society of Heating, Ventilating and Air-Conditioning Engineers (ASHARE). Among them, PMV represents the relationship between the thermal comfort of the human body and the human body's activity level, clothing status, air temperature, humidity, radiation temperature, and wind speed, which are six thermal environment elements, and PPD represents the percentage of dissatisfaction with the thermal environment. PPD is derived from PMV. Among them, the calculation method of PMV is as follows:

PMV＝(0.028+0.3033e ^-0.036M )×{(MW)-3.05[5.733-0.000699(MW)-Pa]

-0.42[(MW)-58.15]-0.0173M(5.867-Pa)-0.0014M(34-T _a ) (1)

-3.96×10 ⁸ ×fcl[(T _cl +273) ⁴ -(T _mrt +273) ⁴ ]-fcl×h _c (T _cl -T _a )}

Among them, Tcl and hc are the convective heat transfer coefficient and the surface temperature of the clothes, respectively, and their calculation methods are as follows:

_Tcl = 35.7-0.028 (MW)

-0.155I _cl {3.96×10 ^-8 ×fcl[(T _cl +273) ⁴ -(T _mrt +273) ⁴ ]-fcl×h _c (T _cl -T _a )} (2)

In the above formulas, M is the metabolic rate, which corresponds to the degree of human activity; W is the work done by the human body, usually zero; Ta is the air temperature; Tmrt is the radiation temperature; Vair is the wind speed; Pa is the water vapor in the surrounding air Partial pressure, which can be calculated from air humidity and temperature: Pa=air humidity×exp(16.6536-4030.183/(Ta+235)); Icl is the thermal resistance from the skin to the outer surface of the human body, corresponding to the clothing condition of the human body; fcl It is a constant related to the clothing condition of the human body, and its value can be set to 1.05, 1.1 or 1.15 according to the different clothing conditions of the human body.

Therefore, by obtaining six parameters such as human activity level, clothing condition, air temperature, humidity, radiant temperature and wind speed, the value of human thermal comfort index (PMV) can be finally calculated.

As shown in Figure 1, according to the PMV index system established by the above equation, the PMV value is divided into 7 grades according to the human thermal sensation. When the PMV value is zero, the human body will be in the best thermal comfort state. Therefore, an HVAC system controller that can control indoor thermal comfort can be designed according to the above theory.

In US Patent Application No. 11/048,740, entitled "Air Conditioning System and Method of Controlling It," a thermal comfort index based air conditioning system is disclosed. However, it only considers the air temperature, the degree of human activity and the amount of clothing, and does not fully consider the six parameters related to the thermal comfort of the human body. However, the PMV value calculated by considering only some parameters is inaccurate, and thus cannot provide the highest comfort for the user. Similarly, the Chinese patent application with the patent number 02136099.5 and titled "Thermal Comfort Fuzzy Control Air Conditioner" only considers four parameters such as air temperature, humidity, radiation temperature and wind speed, making it difficult to precisely control the thermal comfort.

In addition, Chinese patent application No. 200410020153.9 titled "Comfort Index Control Method for Air Conditioning System" and Chinese patent No. 94216957.3 titled "Intelligent Air Conditioner Controller Based on Neural Network and Fuzzy Logic" disclose A method and device for controlling an air conditioning system according to six parameters related to human thermal comfort. However, the disadvantages of the above patents or patent applications are that multiple sensors are used to measure air temperature, humidity, radiation temperature, wind speed and human activity status, etc., and the cost is relatively high. The dressing situation is predicted by the system itself, and the user cannot make a choice, and the flexibility is low. At the same time, because the thermal comfort index PMV value is only controlled within a certain range, and no energy-saving strategy is designed, it is impossible to further reduce system energy consumption.

Contents of the invention

One object of the present invention is to address the defects and deficiencies of the prior art, and provide an HVAC system control method that divides the HVAC system and fans into three operating stages by adopting an energy-saving thermal comfort control method to achieve comfort and energy-saving effects device.

Another object of the present invention is to provide a method to control the heating, ventilation and air-conditioning system through a neural network control algorithm according to six parameters related to human thermal comfort (air temperature, humidity, wind speed, radiant temperature, user activity level and clothing status). A lower cost (only temperature and humidity sensor) controller for control.

Yet another object of the present invention is to provide a HVAC system controller that can be set by the user's activity level and clothing status, thereby improving the accuracy and flexibility of thermal comfort index calculation.

Another object of the present invention is to provide a HVAC system controller that can calculate wind speed according to fan characteristics and input power, thereby avoiding the use of wind speed sensors and reducing the cost of the controller.

Another object of the present invention is to provide a HVAC system controller designed with a DC drive circuit and an AC drive circuit, so that the cooling/heating capacity and DC, DC, The speed of the AC fan can be adjusted separately or simultaneously, that is, one or multiple DC and AC devices can be controlled simultaneously, which ensures the versatility and flexibility of the controller.

In order to achieve the above object, the present invention proposes an energy-saving thermal comfort controller suitable for HVAC systems, including: a sensor module, a multiplex switch module, a central processing unit and a driver; wherein the sensor module is configured For detecting air temperature and humidity; the multi-way selection switch module is configured to set the user's activity level and user's clothing condition, and transmit the user's activity level and user's clothing condition data to the The central processing unit; the central processing unit is configured to calculate the actual value of the thermal comfort index PMV according to the air temperature, humidity, wind speed, radiation temperature, user activity level, and user clothing condition value, and calculate the PMV actual value and a PMV set value, thereby outputting a control signal according to the difference; and the driver is configured to receive the control signal from the central processing unit, and drive the heater according to the control signal A ventilation and air-conditioning system, wherein the HVAC system has multiple operating stages, and the multiple operating stages have different PMV setting values respectively, and the multiple operating stages include: a rapid cooling/heating stage, when When the running time of the HVAC system is less than or equal to the preset time T1, it works in the rapid cooling/heating stage; in the comfortable running stage, when the running time of the HVAC system is longer than the preset time T1 When the preset time T1 is less than or equal to the preset time T2, it works in the comfortable operation stage; and the energy-saving operation stage, which further includes an adjustment stage and a stabilization stage, when the HVAC system is running When the time is greater than the preset time T2 and less than or equal to the preset time T3, it works in the adjustment stage. When the running time of the HVAC system is greater than the preset time After T3, the stable stage is entered, and the controller controls the HVAC system to continue running in the stable stage until it stops running.

In order to achieve the above objects, the present invention also proposes a method for controlling an HVAC system to achieve optimal thermal comfort and minimum energy consumption, comprising the following steps: detecting air temperature and humidity; setting user activity level and user Clothing condition; calculate wind speed; calculate thermal comfort index PMV actual value according to air temperature, humidity, wind speed, radiation temperature and user activity level and clothing condition value, wherein the radiation temperature is set to be equal to the air temperature; according to the determining the thermal comfort index PMV setpoint for a plurality of operating phases of the HVAC system; calculating a difference between the PMV actual value and the PMV setpoint; and operating the HVAC system based on the difference The unit is adjusted to achieve the best thermal comfort and the lowest energy consumption, wherein the multiple operating phases include: a rapid cooling/heating phase, when the HVAC system runs for less than or equal to the preset When the time T1 is set, it works in the fast cooling/heating stage; in the comfort running stage, when the running time of the HVAC system is greater than the preset time T1 and less than or equal to the preset Time T2, it works in the comfortable operation stage; and the energy-saving operation stage, which further includes an adjustment stage and a stabilization stage, when the running time of the HVAC system is greater than the preset time T2 and less than or equal to When the preset time T3 has been set, it works in the adjustment stage, and when the running time of the HVAC system is greater than the preset time T3, it enters the stable stage to control the HVAC system Continue to operate in the stable phase until stopped.

Description of drawings

Fig. 1 is a schematic diagram of the comfort range of human body temperature sensation according to PMV;

Fig. 2 is the block diagram of the hardware structure of the controller according to the preferred embodiment of the present invention;

3 is a schematic diagram of a sensor module according to a preferred embodiment of the present invention;

4 is a schematic diagram of a multiplex switch module according to a preferred embodiment of the present invention;

5 is a schematic diagram of central processing unit data processing according to a preferred embodiment of the present invention;

Fig. 6 is a DC drive circuit diagram according to a preferred embodiment of the present invention;

Fig. 7 is an AC drive circuit diagram according to a preferred embodiment of the present invention;

Fig. 8 is a block diagram of a neural network control algorithm according to a preferred embodiment of the present invention;

Fig. 9 is a flowchart of a control method according to a preferred embodiment of the present invention.

Detailed ways

As mentioned above, the energy-saving thermal comfort controller of the present invention calculates the thermal comfort index PMV based on six parameters (air temperature, air humidity, wind speed, radiant temperature, user activity level and clothing condition). As shown in Figure 1, the value range of PMV index is -3~3. Under normal circumstances, when the PMV value is zero, the thermal environment is the most comfortable, if it is greater than zero, it will be hotter, and if it is less than zero, it will be colder. The controller of the present invention adopts an energy-saving control method, divides the HVAC system into three operation stages of rapid cooling/heating, comfortable operation, and energy-saving operation, and provides corresponding comfort index PMV set values. After obtaining the PMV actual value, the controller compares the PMV actual value with the set value, and according to the deviation between the set value and the actual value, generates a control signal through a neural network control algorithm to control the adjustment of the DC drive and the AC drive respectively. output power to change the indoor thermal environment. By performing the above operations cyclically, the PMV value is controlled near the set value, so as to provide the best thermal comfort under the condition of energy saving.

Preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein the preferred embodiments shown in the drawings are only exemplary.

Fig. 2 is a block diagram of the hardware structure of the controller according to the preferred embodiment of the present invention. The main components of the controller include a sensor module 10 , a multiplex switch module 20 , a central processing unit 30 , a DC driver 40 and an AC driver 50 . The entire controller is supplied with input power by direct current 1 .

Fig. 3 is a schematic structural diagram of a sensor module 10 according to a preferred embodiment of the present invention, including a temperature sensor 110 and a humidity sensor 120, whose power is provided by a central processing unit 30 through a power supply line 1000, and through a temperature data signal line 1100 and a humidity data signal Line 1200 outputs a signal to the central processing unit 30 . In addition, as required, there may be multiple temperature sensors and humidity sensors. In this embodiment, only one temperature sensor and one humidity sensor are preferably used.

Fig. 4 is the structure schematic diagram of the multi-way selection switch module 20 according to the preferred embodiment of the present invention, comprises two three-stage type multi-way selection switches 210 and 220, and its power supply is provided by central processing unit 30 through power supply line 1000, and through signal Lines 2100 and 2200 output signals to the central processing unit 30 . By switching the switches 211 and 221 on the multi-way selector switches 210 and 220 respectively in three different positions, the power supply line 1000 can be connected to different signal lines 2110-2130 and 2210-2230. Multiplex switch 1 210 connects with appropriate signal lines 2110-2130 through switch 211, and then outputs the user's clothing amount data through data signal line 2100. The second multiplex switch 220 connects with appropriate signal lines 2210-2230 through the switch 221, and then outputs user activity data through the data signal line 2200. Thus, the data on the amount of clothing worn and the amount of activity of the user are transmitted to the central processing unit 30 . Here, besides the multi-way selector switch, various devices well known to those skilled in the art can be used to realize the composition of the multi-way selector switch module, as long as it can realize the gear selection purpose required by the present invention. Descriptions of other optional implementation manners are omitted here.

Fig. 5 is a schematic diagram of data processing by a central processing unit according to a preferred embodiment of the present invention. The central processing unit 30 cyclically executes each operation command, and in each cycle, the following operations are performed: read the temperature data signal line 1100 and the humidity data passing through the sensor module 10 through the first serial interface 311 and the second serial interface 312 The temperature and humidity signals transmitted by the data signal line 1200 are converted from analog to digital by the first A/D converter 321 and the second A/D converter 322 , and then the corresponding digital signals are stored in the memory 330 . Through the first parallel interface 313 and the second parallel interface 314, the user's clothing amount data and activity data transmitted through the data signal lines 2100 and 2200 of the multiplex switch module 20 are read, and the data are directly stored in the memory 330 . After the memory 330 receives a complete set of input data (including temperature, humidity, user clothing and user activity), the latest data is transmitted to the arithmetic logic unit 340 for calculating the actual value of PMV.

In addition, the arithmetic logic unit 340 is also provided with a counter 350. The purpose of the counter 350 is: (1) as a timer for recording the working time, in order to compare the current working time of the system with the set time, and determine on this basis PMV setting value (will be described in detail below); (2) Control the pulse width (duty cycle) of pulse width modulation, so as to control the speed of the DC motor by adjusting the pulse width.

6 is a circuit diagram of a DC driver 40 according to a preferred embodiment of the present invention, which mainly includes a buffer 410 and a high-power N-type bipolar transistor 420, and receives pulse width modulation from a pulse width modulation signal bus 3100 of a central processing unit 30. signal as input. The pulse width modulation signal bus 3100 provides 5 volts to the buffer 410 through the wire 3120, and the pulse width modulation signal is input to the buffer 410 through the signal line 3110, and the buffered pulse width modulation signal controls the N-type bipolar through the wire 3130 The transistor 420 operates to regulate the input power provided by the DC 2 to the DC motor 60 . Therefore, through the designed DC drive 40, DC drive equipment such as a DC fan can be controlled.

7 is a structural diagram of an AC driver 50 according to a preferred embodiment of the present invention, which mainly includes a multiplexer 510, a transformer 530, a first relay 521 to a fifth relay 525, and receives four bits from the central processing unit 30. A four-bit binary signal from the binary signal bus 3200 is used as input. The alternating current 3 is converted into five sets of different voltages by the transformer 530 , and output at the first interface 531 to the fifth interface 535 respectively. The AC motor 70 is connected to one of the interfaces through a wire 5100 . The four-bit binary signal selects and activates the appropriate relay among the first relay 521 to the fifth relay 525 through the multiplexer 510 , and makes the voltage of the interface connected thereto provided to the AC motor 70 through the wire 5100 . Therefore, through the designed AC drive 50, the output voltage can be changed to control various types of AC drive equipment such as AC fans and compressors.

The specific circuits of the DC driver and the AC driver described here are only exemplary, and it is obvious to those skilled in the art that other forms of driving circuits that can change power according to pulse width modulation signals or binary signals can be used.

The control method of the present invention will be described below. According to the present invention, the method for controlling the HVAC system to provide the best thermal comfort is mainly carried out through three steps, namely: (1) acquiring parameter data; (2) data processing and calculation; (3) outputting control signals . The three steps will be described below in conjunction with the preferred embodiments shown in the accompanying drawings.

(1) Get parameter data

In the present invention, the thermal comfort index PMV is calculated according to six parameters such as air temperature, air humidity, wind speed, radiation temperature, user's activity level and clothing condition, so as to control the thermal comfort. Therefore, it is necessary to first determine the specific values of the six parameters. The air temperature and air humidity are respectively measured by the temperature sensor 110 and the humidity sensor 120 in the sensor module 10, and the radiation temperature is set to be consistent with the measured air temperature. The measured air temperature, radiation temperature and air humidity data are transmitted to the central processing unit 30 through the temperature data signal line 1100 and the humidity data signal line 1200 .

Data on the user's activity level and attire are obtained from the multi-way switch module 20. As described above, the user sets his activity level by switching the multi-way switch one 210 on three different positions to "weak", "Medium" and "Strong" three gears. When the user selects the "weak" gear, the user's metabolic rate used to indicate the degree of activity is set to 0.8Met (equivalent to 46W/m ² ); when the user selects the "medium" gear, the metabolic rate is set to 1.2Met (equivalent to 70W/m2 ² ); when the "strong" file is selected, the metabolic rate is set to 2.2Met (equivalent to 125W/m ² ). By default, user activity is set to "medium". In addition, the user sets his clothing status to three levels of "weak", "medium" and "strong" by switching the multi-way selector switch 2 220 on three different gears. When the user selects the "weak" gear, the user's clothing volume is set to 0.5clo (equivalent to 0.0755m ² K/W); when the "medium" gear is selected, the user's clothing volume is set to 0.8clo (equivalent to 0.124m ² K/W ); when the "strong" file is selected, the user's clothing volume is set to 1.1clo (equivalent to 0.1705m ² K/W). By default, the user's clothing status is set to "medium". After the user selects, the multi-way selection switch module 20 transmits the data of the user's clothing amount and activity amount in the form of digital signals to the central processing unit 30 through the data signal lines 2100 and 2200 respectively.

The wind speed is calculated by the central processing unit according to the fan characteristics and input power of the HVAC system. For a specific type of fan, there is a clear correspondence between its input power, rotational speed and outlet wind speed. For the user, the wind speed sensed by the user is proportional to the distance from the fan. Therefore, according to the cooling/heating capacity of the HVAC system, the size of the indoor space where it is applied can be known, and the average distance between the user and the fan can be further calculated. Based on this, a relational function between the fan input power and the user-induced wind speed can be obtained by fitting, and stored in the central processing unit 30 . The relationship function can be fitted with a polynomial of degree 3 or degree 2 for the input power. For example, for a DC fan with a rated input power of 12W, when the distance between the user and the fan is 1 meter, the relational function expression can be fitted by the least square method as:

V _air ＝-0.7505×P ² +16.8504×P-90.8250,

Among them, V _air is the wind speed, and P is the fan power. When the distance between the user and the fan is 1.5 meters, the relational function expression can be fitted as:

V _air =-0.1201×P ² +2.9950×P-16.4640,

When the distance between the user and the fan is 2 meters, the relational function expression can be fitted as:

V _air =-0.2787×P ² +6.3470×P-34.4057.

When the energy-saving thermal comfort controller starts to work, the central processing unit can calculate the wind speed value through the obtained functional expression according to the initial power of the fan, so as to obtain the control signal for controlling the input power of the fan according to the six parameters. Correspondingly, since the control signal controls the input power of the fan, the input power of the fan can be determined when the control signal is known, so as to obtain the wind speed value for the next control. Thus, the wind speed can be calculated according to the characteristics of the wind turbine and the input power, thereby eliminating the need to use a wind speed sensor and reducing the cost of the controller.

(2) Data processing and calculation

As mentioned above, the central processing unit 30 receives the temperature and humidity data from the sensor module 10 , performs A/D conversion and stores them in the memory 330 . The user's clothing amount data and activity amount data from the multiplex switch module 20 are digital signals and thus are directly stored in the memory. The memory 330 transmits the received air temperature, humidity, radiation temperature (which is set equal to the air temperature), user's clothing amount, and user's activity amount data to the arithmetic logic part 340 . The arithmetic logic unit 340 also receives count information transmitted from the counter 350 at the same time.

The arithmetic logic unit 340 first obtains the wind speed according to the fan characteristics and input power of the HVAC system, and then executes the thermal comfort index PMV algorithm according to the remaining parameter values obtained from the memory 330 to obtain the actual value of PMV, and displays the obtained value on the LED display unit 80. Then, the arithmetic logic unit 340 executes the energy-saving thermal comfort control method, and determines the working phase according to the working time of the system, so as to determine the PMV setting value.

The energy-saving thermal comfort control method of the present invention divides the HVAC system into different operating stages, and each operating stage has different PMV set values. By judging which operating stage the system is currently in according to the system running time, the PMV set value can be obtained, so as to control the HVAC system to work in the required stage.

In a preferred embodiment of the present invention, the HVAC system is divided into three operating stages: fast cooling/heating, comfortable operation and energy-saving operation, and the specified operating times T1, T2 and T3 of each stage have been preset. By comparing the current running time of the system with T1, T2 and T3, it can be judged which stage the system should work in, so as to obtain the PMV set value. Figure 9(b) shows the above workflow, as shown in the figure, when the controller starts to work, the arithmetic logic unit 340 first reads the user instruction to determine which mode of operation the air-conditioning system will work in, cooling or heating, Then compare the current working time t of the controller with the specified running time of each stage, so as to determine the working stage of the controller and obtain the PMV set value.

The following uses refrigeration as an example to illustrate: firstly, it is judged whether t is greater than T1, if not, then the PMV setting value is -0.5, and if so, it is judged whether t is greater than T2. If t is less than or equal to T2, the PMV setting value is 0; if t is greater than T2, it is judged whether t is greater than T3. If t is less than or equal to T3, the PMV set value increases from 0 to 0.5 within the time period T3-T2, and if t is greater than T3, the PMV set value is 0.5. In a preferred embodiment of the present invention, T1, T2 and T3 can be set to 0.5 hours, 2 hours and 4 hours respectively, namely within 0.5 hours after the system is turned on, the PMV set value is -0.5; During the period between 0.5 hours and less than or equal to 2 hours, the PMV setting value is 0; when the power-on time is greater than 2 hours and less than or equal to 4 hours, the PMV setting value gradually increases from 0 to 0.5, and when the power-on time is greater than 4 hours, PMV The setpoint is kept at 0.5 continuously.

After the PMV set value is determined according to the above thermal comfort control method, the arithmetic logic unit 340 calculates the difference between the PMV set value and the actual value, and uses the neural network control algorithm to obtain the HVAC system based on the above difference. output signal. Different from the traditional switch control or PID control, the energy-saving thermal comfort controller adopts the neural network control algorithm, thus ensuring a good self-adaptive self-learning ability, and it can still be used under different PMV set values and ambient temperature and humidity disturbances. Maintain good control performance.

和

网络权重为

和神经网络输出节点的输入和输出变量分别为

和u，网络权重为

其中m＝1，2……4。Fig. 8 is a block diagram of a neural network control algorithm according to a preferred embodiment of the present invention. In practical applications, the neural network can adopt a 2-layer or 3-layer network structure. In the following, a neural network with 3-layer architecture with 4 intermediate nodes is taken as an example to describe the neural network control algorithm in detail. As shown in Figure 8, the input and output variables of each intermediate node (that is, each neuron) are respectively

and

The network weight is

and The input and output variables of the neural network output node are respectively

and u, the network weight is

where m=1, 2...4.

作为神经网络的输入变量，并根据下式，计算中间各节点的输入：First, take the difference e between the PMV setpoint and actual value, and the rate of change of that difference

As the input variable of the neural network, and according to the following formula, calculate the input of each node in the middle:

${\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; e</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="w\; m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">w</figure-callout></span><figure-callout\; id="2"\; label="w\; m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}^{\mathrm{<span\; class="notranslate">m</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}\stackrel{<span\; class="notranslate">\; \&Center\; Dot;</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

Where θ is the offset value. The output of each intermediate node is calculated using the Unipolar Sigmoid function:

${\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

Similarly, the input and output of the output node of the neural network are calculated according to the following formulas:

${\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; 4</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

$\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; exp</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; I</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

The output u serves as a control signal for controlling the HVAC system. Since the output variable is controlled between 0 and 1, there is no need to limit the output control signal. For network weights, the backpropagation algorithm can be used to update:

$\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&eta;</span>}\frac{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; E.</span>}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&eta;</span>}\frac{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; E.</span>}}{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}}{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; u</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; u</span>}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

In the formula, E is the error function, defined as

$\mathrm{<span\; class="notranslate">\; E.</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; e</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$

Since the thermal comfort controller is in cooling mode, is a negative value, so the above formula can be further simplified as:

$\mathrm{<span\; class="notranslate">\; \&Delta;</span>}{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}^{<span\; class="notranslate">\; *</span>}\frac{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; E.</span>}}{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}}\frac{<span\; class="notranslate">\; \&PartialD;</span>\mathrm{<span\; class="notranslate">\; u</span>}}{<span\; class="notranslate">\; \&PartialD;</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; ij</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; )</span>$

In the formula, η ^* is the weight adjustment speed. Based on this, the network weights of the output layer and the middle layer are updated according to the following formula:

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">m</figure-callout></span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; SV</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 11</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; SV</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; e</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 12</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="w\; m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">w</figure-callout></span><figure-callout\; id="2"\; label="w\; m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}^{\mathrm{<span\; class="notranslate">m</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="2"\; label="w\; m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">w</figure-callout></span><figure-callout\; id="2"\; label="w\; m"\; filenames="H061C8692320060919E000051.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}^{\mathrm{<span\; class="notranslate">m</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&eta;</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}<span\; class="notranslate">\; \_</span>\mathrm{<span\; class="notranslate">\; SV</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; PMVs</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; v</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span>\stackrel{<span\; class="notranslate">\; \&Center\; Dot;</span>}{\mathrm{<span\; class="notranslate">\; e</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 13</span>}<span\; class="notranslate">\; )</span>$

Wherein, m=1, 2...4.

In summary, the implementation process of the neural network control algorithm can be briefly described as follows: First, the difference between the PMV set value and the actual value, and the rate of change of the difference, are used as the input variables of the designed multi-layer neural network ; Then update the input and output variables and weights of each neuron according to the backpropagation algorithm; finally calculate and output the control signal.

(3) Output control signal

After completing the data processing and calculation, the arithmetic logic unit 340 outputs two kinds of control signals: (1) digital signal, which is used to control the pulse width modulation processor 360 to generate the required pulse width modulation signal, and the pulse width modulation signal is compatible with 5 volts The voltage 370 is delivered to the DC driver 40 through the pulse width modulation signal bus 3100 together to control the DC driver 40; (2) four-bit binary signals are directly delivered to the AC driver 50 through the four-bit binary signal bus 3200 to control the AC driver 40. The driver 50 performs control. Thus, the DC driver 40 and the AC driver 50 respectively or simultaneously adjust the cooling/heating capacity and fan speed of HVAC systems such as air conditioners and air conditioners according to the received control signals. When adjusting the fan, since the continuous speed regulation of the DC fan and the multi-stage speed regulation of the AC fan can be realized through the DC and AC drivers, compared with the conventional 3-speed fan adjustment, the controller of the present invention can Provides better thermal comfort.

Fig. 9 is a flowchart of a control method according to a preferred embodiment of the present invention, which generally shows the flow of the control method of the present invention. As shown in Figure 9(a), after the controller starts to work, the sensor first reads the air temperature (the radiation temperature is set to be consistent with the air temperature) and the humidity value (step S10), and the multi-way switch obtains user input activities and clothing values (step S20), and calculate the wind speed value (step S30) by fan input power; after obtaining the above six parameters, calculate the PMV actual value (step S40) by the central processing unit; then calculate the PMV actual value and According to the deviation of the set value (step S50), the control signal is obtained by executing the neural network control algorithm according to the obtained deviation (step S60), so as to control the DC and AC drive equipment (step S70), so that the two generate DC and AC output respectively, through The output is used to control the cooling/heating capacity and fan speed of the HVAC system, so as to reasonably adjust the indoor thermal comfort and provide users with satisfactory thermal comfort.

Apparently, the preferred embodiments described above are only illustrative descriptions of the present invention, and should not be regarded as limitations to the present invention. It should be recognized that the above-described electronic components and their connections are only exemplary, and those skilled in the art can make various modifications and replacements without departing from the spirit of the present invention. In addition, the above-mentioned control methods are only As a preferred solution, those skilled in the art can properly select and optimize the steps in the method according to specific conditions. Therefore, any changes and modifications that do not depart from the essence of the present invention shall fall within the protection scope of the present invention. The protection scope of the present invention is defined by the appended claims.

Claims (15)

1. A controller for an hvac system comprising a sensor module, a multiplexer module, a central processing unit and a driver, wherein:

the sensor module is configured to detect air temperature and humidity;

the multi-way selection switch module is configured to be used for setting the activity degree of the user and the clothes wearing condition of the user and respectively transmitting the activity degree of the user and the clothes wearing condition data of the user to the central processing unit through two data signal lines;

the central processing unit is configured to calculate a thermal comfort index PMV actual value based on the air temperature, humidity, wind speed, radiation temperature, and user activity level and user dressing condition values, and calculate a difference between the PMV actual value and a PMV set value, thereby outputting a control signal based on the difference; and

the driver is configured to receive a control signal from the central processing unit and drive the HVAC system according to the control signal,

wherein the HVAC system has a plurality of operating phases, the plurality of operating phases having different PMV set values, respectively, the plurality of operating phases including:

a rapid cooling/heating stage, when the running time of the heating, ventilating and air conditioning system is less than or equal to the preset time T1, the heating, ventilating and air conditioning system works in the rapid cooling/heating stage;

a comfortable operation stage, when the operation time of the heating, ventilating and air conditioning system is greater than the preset time T1 and less than or equal to the preset time T2, the heating, ventilating and air conditioning system works in the comfortable operation stage; and

and the energy-saving operation stage further comprises an adjusting stage and a stabilizing stage, when the operation time of the HVAC system is greater than the preset time T2 and less than or equal to the preset time T3, the HVAC system operates in the adjusting stage, and when the operation time of the HVAC system is greater than the preset time T3, the HVAC system enters the stabilizing stage, and the controller controls the HVAC system to continuously operate in the stabilizing stage until the HVAC system stops operating.

2. The controller of claim 1, wherein the central processing unit determines the PMV setpoint according to the plurality of operating phases:

when the heating, ventilating and air conditioning system works in the rapid cooling/heating stage, the PMV set values are-0.5 and 0.5 respectively under the cooling and heating conditions;

when the heating, ventilating and air conditioning system works in the comfortable operation stage, the PMV set value is 0; and

when the heating, ventilation and air conditioning system works in the adjusting stage, the PMV set value is gradually increased from 0 to 0.5 under the condition of cooling, and is gradually decreased from 0 to-0.5 under the condition of heating, and after the heating, ventilation and air conditioning system enters the stabilizing stage, the PMV set value is respectively 0.5 and-0.5 under the conditions of cooling and heating.

3. The controller of claim 1, wherein the central processing unit derives the control signal using a neural network control algorithm having the steps of:

(a) the difference e between the PMV setpoint and the actual value, and the rate of change of the difference

As input variables of the multi-layer neural network;

(b) updating the input and output variables and the weight of each neuron according to a back propagation algorithm; and

(c) and calculating and outputting the control signal.

4. The controller of claim 3, wherein the input variables of each neuron in step (b)And an output variable

The calculation method of (2) is as follows:

<math><mrow><msubsup><mi>I</mi><mi>m</mi><mn>1</mn></msubsup><mo>=</mo><msubsup><mi>w</mi><mrow><mi>m</mi><mn>1</mn></mrow><mn>1</mn></msubsup><mi>e</mi><mo>+</mo><msubsup><mi>w</mi><mrow><mi>m</mi><mn>2</mn></mrow><mn>1</mn></msubsup><mover><mi>e</mi><mo>&CenterDot;</mo></mover><mo>+</mo><msubsup><mi>&theta;</mi><mi>m</mi><mn>1</mn></msubsup></mrow></math>

$v_m^1=\frac{1}{1+\exp (-I_m^1)}$

wherein m is 1, 2 … … 4,

the network weight w is updated as follows:

<math><mrow><mi>&Delta;</mi><msub><mi>w</mi><mi>ij</mi></msub><mo>=</mo><mo>-</mo><mi>&eta;</mi><mfrac><mrow><mo>&PartialD;</mo><mi>E</mi></mrow><mrow><mo>&PartialD;</mo><msub><mi>w</mi><mi>ij</mi></msub></mrow></mfrac><mo>=</mo><mo>-</mo><mi>&eta;</mi><mfrac><mrow><mo>&PartialD;</mo><mi>E</mi></mrow><mrow><mo>&PartialD;</mo><mi>PMV</mi></mrow></mfrac><mfrac><mrow><mo>&PartialD;</mo><mi>PMV</mi></mrow><mrow><mo>&PartialD;</mo><mi>u</mi></mrow></mfrac><mfrac><mrow><mo>&PartialD;</mo><mi>u</mi></mrow><mrow><mo>&PartialD;</mo><msub><mi>w</mi><mi>ij</mi></msub></mrow></mfrac></mrow></math>

where θ is the offset value, E is the error function, and u is the output control signal.

5. The controller of claim 4, wherein the control signal u in step (c) is calculated as follows:

$u=\frac{1}{1+\exp (-I_1^2)},$

wherein, <math><mrow><msubsup><mi>I</mi><mn>1</mn><mn>2</mn></msubsup><mo>=</mo><munderover><mi>&Sigma;</mi><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mn>4</mn></munderover><msubsup><mi>w</mi><mrow><mn>1</mn><mi>i</mi></mrow><mn>2</mn></msubsup><msub><mi>v</mi><mi>i</mi></msub><mo>+</mo><msubsup><mi>&theta;</mi><mn>1</mn><mn>2</mn></msubsup><mo>.</mo></mrow></math>

6. the controller of claim 1, wherein the sensor module comprises a temperature sensor and a humidity sensor.

7. The controller of claim 1, wherein the multi-way selection switch module comprises selection switches for setting the degree of user activity and the dressing condition of the user, respectively.

8. The controller of claim 7, wherein the user activity level is expressed as a metabolism rate divided into 0.8Met, 1.2Met and 2.2Met corresponding to a weak, medium and strong gear of the selection switch for setting the user activity level, respectively.

9. The controller of claim 7, wherein the user's dressing conditions are classified into 0.5clo, 0.8clo and 1.1clo, corresponding to the weak, medium and strong stages of the selection switch for setting the user's dressing conditions, respectively.

10. The controller of claim 1, wherein the driver comprises a dc driver including N-type bipolar transistors generating different output powers according to a pulse width modulation signal as a control signal input to the dc driver to control a dc device.

11. The controller of claim 1, wherein the driver comprises an ac driver connected to different transformer coils using multiplexers according to an input binary signal as a control signal to vary output voltage and power to control the ac device.

12. The controller of claim 1, further comprising a display unit for displaying the PMV actual value.

13. A method of controlling an hvac system for achieving optimal thermal comfort, comprising the steps of:

detecting the temperature and humidity of air;

setting the activity degree and the dressing condition of the user;

calculating the wind speed;

calculating a thermal comfort index (PMV) actual value according to an air temperature, a humidity, a wind speed, a radiation temperature, a user activity degree and a dressing condition value, wherein the radiation temperature is set to be equal to the air temperature;

determining the thermal comfort index PMV set value according to a plurality of operation stages of the heating, ventilating and air conditioning system;

calculating a difference between the PMV actual value and a PMV set value; and

adjusting the working units of the HVAC system according to the difference to achieve optimal thermal comfort and lowest energy consumption,

wherein the plurality of operational phases comprises:

a rapid cooling/heating stage, when the running time of the heating, ventilating and air conditioning system is less than or equal to the preset time T1, the heating, ventilating and air conditioning system works in the rapid cooling/heating stage;

a comfortable operation stage, when the operation time of the heating, ventilating and air conditioning system is greater than the preset time T1 and less than or equal to the preset time T2, the heating, ventilating and air conditioning system works in the comfortable operation stage; and

and the energy-saving operation stage further comprises an adjusting stage and a stabilizing stage, when the operation time of the heating, ventilating and air conditioning system is greater than the preset time T2 and less than or equal to the preset time T3, the operation is carried out in the adjusting stage, and when the operation time of the heating, ventilating and air conditioning system is greater than the preset time T3, the stabilizing stage is carried out, and the heating, ventilating and air conditioning system is controlled to continuously operate in the stabilizing stage until the operation is stopped.

14. The method of claim 13, wherein determining the thermal comfort index PMV setpoint based on the plurality of operating phases comprises:

when the heating, ventilation and air conditioning system is in the rapid cooling/heating stage, the PMV set values are-0.5 and 0.5 respectively under the cooling and heating conditions;

when the heating, ventilating and air conditioning system is in the comfortable operation stage, the PMV set value is 0; and

when the heating, ventilating and air conditioning system is in the energy-saving operation stage, the PMV set value is firstly changed within a certain time period and then fixed at a certain fixed value, namely, within the time period, the PMV set value is gradually increased from 0 to 0.5 under the condition of cooling, and is gradually decreased from 0 to-0.5 under the condition of heating, and after the time period, the PMV set values are respectively 0.5 and-0.5 under the conditions of cooling and heating.

15. The method of claim 13, wherein the step of adjusting the operating units of the hvac system based on the difference is accomplished by a neural network control algorithm having the steps of:

(a) taking the difference between the PMV set value and the actual value and the change rate of the difference as input variables of the multilayer neural network;

(b) updating the input and output variables and the weight of each neuron according to a back propagation algorithm; and

(c) and calculating and outputting the control signal.

Images