Central air conditioner charging device and system based on comfort level

Abstract

The invention discloses a central air conditioner charging device and system based on comfort. The device comprises a control unit, a sensing detection unit, a user interface unit and a working condition adjusting unit. Establishing a first neural network and a second neural network to model the cooling capacity characteristic of the end unit and the cooling characteristics of the rooms in different directions, and predicting the cooling capacity of the end unit and the cooling capacity of the rooms under the actual working condition based on the two models; and on the basis of the difference of the cold consumption characteristics of the rooms in different directions, the predicted actual cold consumption is corrected on the basis of a working condition conversion coefficient and a temperature comfort coefficient, and then the cost of the central air conditioner is shared. The method can dynamically reflect the change of the cooling capacity of the tail end unit, and compared with a tail end detection method, the metering precision is improved; through correction, various influence factors such as sunlight and adjacent chamber heat transfer are effectively compensated. According to the method, the cooling capacity of the central air conditioner terminal unit under different working conditions can be converted, fair charging according to the comfort degree is achieved, and economical utilization of the air conditioner is facilitated.

Description

[0001] This application is a divisional application of application No. 202211358529.1, titled "Central air conditioning billing device and system", with a filing date of November 1, 2022. TECHNICAL FIELD

[0002] The present application relates to the field of central air conditioning metering and billing, in particular to a central air conditioning billing device and system based on comfort level. BACKGROUND

[0003] A central air conditioning system is composed of a cold and heat source system and a cold and heat transfer and air conditioning system. In summer, the refrigeration system provides the required cold energy for the air conditioning system to offset the heat load of the indoor environment; in winter, the heating system provides the cold energy for the air conditioning system to offset the heat load of the indoor environment. The central air conditioner supplies cold to each terminal unit through the refrigerant pipeline of the main unit, and the refrigerant can be any of water, air or refrigerant. Taking the water-cooled central air conditioner as an example, it is an air conditioner that supplies cold to different rooms by connecting multiple fan coil terminal units to the main unit through cold water pipes to achieve the purpose of indoor air conditioning.

[0004] The most prominent feature of central air conditioning is to provide a comfortable working and living environment. With the rapid development of air conditioning demand in large public buildings in China, more and more office buildings, shopping malls, hotel apartments and other buildings have begun to install central air conditioning systems.

[0005] In the new century, mankind is facing two major problems of energy and environment, namely energy shortage and environmental degradation. The construction industry, industrial production and transportation have become China's three major energy consumption industries, accounting for about 30% of the country's energy consumption. The energy consumption of air conditioning and heating systems accounts for 50%-60% of the total building energy consumption. Energy-saving measures for central air conditioning in public buildings can effectively reduce building energy consumption and achieve sustainable development. However, the billing method for central air conditioning in many places still uses the traditional and simple area allocation method; this method is simple and convenient, but under this billing method, most people do not consider whether the air conditioner is used energy-efficiently, resulting in the phenomenon of some people leaving the air conditioner on, leading to energy waste.

[0006] If the energy metering method is used, pay as much as you use, which can awaken people's energy-saving awareness. Adopting a reasonable billing method can change consumers' energy use habits, therefore, it is of great significance to choose and adopt a central air conditioning billing method reasonably.

[0007] The initial charging method for central air conditioning is area allocation, so there is a waste of air conditioning. Therefore, in recent years, a new household billing system based on central air conditioning has been proposed, hoping to reasonably allocate costs to each room user, achieve reasonable billing, and reduce the phenomenon of serious electricity waste and high building energy consumption.

[0008] Among the new technologies for cold metering and cost allocation of central air conditioning, chilled water metering method appears. Chilled water metering method is divided into two kinds. One is to install water meter at the outlet of fan coil, and to meter the chilled water flow in fan coil through water meter. This method simply solves the problem of different cost for different use, but does not consider the chilled water inlet and outlet temperature. The other method is just the opposite, considering that chilled water flow is constant, only the chilled water inlet and outlet temperature difference is metered, and the cost is charged only when the air conditioner is turned on. These methods, including the subsequent improved energy meter method, detect several fixed parameters at the end, and cannot reflect the influence of overall working condition change of central air conditioning system on end cold supply, so it is difficult to reflect the real cold consumption of users.

[0009] Taking water-cooled central air conditioning as an example, another problem that needs attention in traditional cold metering is that the supply and return water temperature difference is much smaller than that of heating, so the measurement accuracy of temperature sensor needs to be higher. Undoubtedly, using high-precision sensors at each end will greatly increase the cost of users; and using ordinary sensors generally has the problem of large sampling fluctuation and inaccurate metering.

[0010] High-cost energy meter is difficult to promote. At present, according to the investigation, the new cold metering and cost allocation system of various central air conditioners is mostly used in engineering projects with indirect charging or equivalent charging, and the time type charging is mostly used in indirect or equivalent charging. The cold equivalent or cumulative consumption of time type metering is the calculated amount under the rated test condition. For example, the basic principle of Chinese patent with publication number CN 100504338C is to calculate the cumulative time of each fan coil speed. Time type charging starts from the proportionality of refrigeration capacity, water temperature difference of fan coil and water volume, and considers that the larger the fan coil speed, the larger the air volume flow, and the larger the cold equivalent. For the fan coil with high, medium and low speed V H , V M , V L , the cold equivalent is:

[0011] Q=K H t H +K M t M +K L t L ,

[0012] Wherein, t H , t M , t L are the opening time (s) of two-way valve under high, medium and low speed, K H , K M , K LThe proportional coefficient (kJ / s) under high, medium and low wind speed. The two-way valve itself is a switch component, by detecting the opening and closing of each two-way valve, so that the cumulative time of two-way valve opening in a certain time and the different speed of wind can be obtained. It can be seen that in the time type measurement method, the cooling capacity of each terminal fan of central air conditioning is calculated by using the cumulative value of the opening time of two-way valve under different wind speed as the basis. The key of this method is how to obtain the K H , K M , K L coefficient? The current method is to use the estimated empirical or theoretical value, or to use the coefficient value measured by the fan coil manufacturer based on the rated conditions such as dry bulb 27℃, wet bulb 19.5℃, chilled water inlet temperature 7℃, and the cooling capacity under each wind speed.

[0013] This method of calculating cooling capacity under dynamic conditions with fixed coefficient can only be used for estimation and cannot be accurately measured.

[0014] Central air conditioning system cooling capacity measurement is the basis for cooling capacity billing, but from measurement to billing, many factors need to be considered. For general buildings, cooling capacity cost cannot be simply calculated by multiplying air conditioning cooling capacity consumption by cooling capacity unit price, and the influence of different positions of air conditioning rooms and inter-room heat transfer also needs to be considered.

[0015] Because the room heat load is related to environmental parameters, solar radiation, maintenance structure heat transfer characteristics and other factors, the heat load of rooms with different orientations such as south and north, and different heights such as top, middle and bottom layers may differ greatly. Therefore, the central cooling system should consider the cooling capacity or unit price of rooms in different positions and orientations. In addition, inter-room heat transfer is also an important factor. In the case of low or vacant use of adjacent rooms, the cooling load of air conditioning rooms will increase. If only the cooling capacity is charged, the cooling capacity cost of rooms with the same area will differ greatly, which is also unfair.

[0016] However, due to the coupling of sunlight, structure and adjacent room and other factors, it is difficult to independently distinguish the influence degree of each factor, and the empirical coefficient method is often used in existing billing methods. For example, Tian Yuchen of Tianjin University in the degree thesis "Research on Related Problems of Heat Metering" combined with the reference specification to calculate the correction coefficient of each household by using statistical method for different building structures, different room types and different orientations. The correction coefficient of rooms in different positions of non-energy-saving residential buildings is calculated, and the correction coefficient of 9 types of rooms is given as 0.55-1.00. This example method can only give a rough guide and cannot accurately determine the cooling consumption relationship of rooms in different directions of specific buildings.

[0017] Therefore, the industry urgently needs a set of devices and systems that can accurately measure the cooling equivalent of the central air conditioning terminal under actual working conditions, correct the cooling consumption of different orientation rooms, and thus reasonably allocate the charges based on the cooling consumption and quantitatively correct the environmental factors, so as to achieve the goal of paying more for more use and paying less for less use, thereby ultimately improving the energy-saving awareness of the majority of users and achieving the purpose of energy saving and environmental protection. SUMMARY

[0018] Therefore, the purpose of the present application is to provide a device and system that can accurately measure the cooling of the terminal unit in the central air conditioning and correct the cooling of each room, so as to achieve reasonable and effective billing based on the actual cooling consumption.

[0019] The technical solution of the present application is to provide a central air conditioning billing device based on comfort, which comprises a control unit, and a sensing detection unit, a user interface unit and a working condition adjustment unit connected to the control unit.

[0020] The sensing detection unit detects the working condition and environmental conditions of the central air conditioning;

[0021] The control unit is configured to:

[0022] First, the cooling equivalent of the central air conditioning terminal unit is modeled,

[0023] The room where the terminal unit is located is the thermal load, the working condition of the central air conditioning system is the input quantity, and the unit time cooling equivalent of the terminal unit in the room is the output quantity. A first neural network is established in the control unit,

[0024] The cooling equivalent of the terminal unit is calculated based on the cooling equivalent of the reference air conditioning unit in the system under the same thermal load condition, and the first data sample is collected under different working conditions and the first neural network is trained;

[0025] Secondly, the central air conditioning cooling rooms are classified according to the orientation type, and a cooling consumption model is established for each type of orientation room,

[0026] The current room temperature and outdoor temperature during cooling, as well as the vector composed of the indoor temperature values in the six directions of the current room, are used as the input quantity, and the unit time cooling equivalent of the current room is used as the output quantity. A second neural network is established in the control unit according to the room,

[0027] The terminal unit in the current room is adaptively controlled, and the room temperature is kept constant in stages. When the working condition is approximately stable, the second data sample of the second neural network under different input conditions is collected, the output quantity in the sample is predicted by the first neural network, and the training is based on the collected data sample;

[0028] Based on the offline established cooling supply and consumption models, the central air conditioner is cost allocated in time periods in online application,

[0029] First, according to the cooling consumption model, the correction coefficient of the current room condition is calculated In the formula, i is the current room number, the room number i = 1 ~ N is any integer, N is the total number of rooms, p i , p i0 is the mapping output of the second neural network when is taken as the input vector, is the current condition of room i, The current room temperature and outdoor temperature are the same as , and the room indoor temperature values in the six directions of the current room are the current room temperature;

[0030] The cost of the i-th room in the d-th time period is calculated as:

[0031] Among them, PF i (t) is the equivalent cooling capacity of the current room based on the first neural network prediction of the end unit cooling capacity per unit time, Q jd is the current period cooling capacity of room j, k j3 is the corresponding coefficient of room j, C d is the total cost of the central air conditioner to be allocated in the current period.

[0032] In another embodiment of the present application, a central air conditioner billing device based on comfort is provided, comprising: a control unit, and a sensing detection unit, a user interface unit and a working condition adjusting unit connected with the control unit;

[0033] The sensing detection unit detects the working condition and environmental conditions of the central air conditioner;

[0034] The control unit is configured to:

[0035] First, the cooling equivalent of the central air conditioner end unit is modeled,

[0036] The room where the end unit is located is taken as the thermal load, the central air conditioner system working condition is taken as the input quantity, and the unit time cooling equivalent of the end unit in the room is taken as the output quantity. A first neural network is established in the control unit,

[0037] The cooling capacity of the reference air conditioning unit in the system under the same thermal load condition is taken as the cooling equivalent of the end unit, and the first data sample is collected under different working conditions and the first neural network is trained;

[0038] Secondly, the central air conditioning cooling rooms are classified by orientation type, and a cooling consumption model is established for each type of orientation room,

[0039] The current room temperature, outdoor temperature, and a vector composed of the indoor temperature values in the six directions of the current room are taken as input quantities, and the cooling consumption per unit time of the current room is taken as an output quantity. A second neural network is established in the control unit according to the room,

[0040] The end unit in the current room is adaptively controlled to keep the room temperature unchanged in stages. When the working condition is approximately stable, the second data sample of the second neural network under different input conditions is collected, the output quantity in the sample is predicted by the first neural network, and training is performed based on the collected data sample;

[0041] Based on the offline established cooling and cooling consumption models, the central air conditioner is charged by time period in online application,

[0042] First, the conversion coefficient of the current room working condition is calculated In the formula, i is the serial number of the current room, s i , s j are the areas of the i-th and j-th rooms, respectively, i and j are any integer in the range of 1 to N, N is the total number of rooms, p i , p j are the respective mapping outputs of the second neural network when the i-th and j-th rooms take as the input vector, is the current working condition of the room i,

[0043] and the comfort coefficient is calculated In the formula, p ic is the mapping output of the second neural network when the i-th room takes as the input vector, The difference between the working conditions is that the current room temperature is a preset comfortable temperature;

[0044] The cost of the i-th room in the d-th time period is calculated as:

[0045] Where k j1 , k j2 are the coefficients corresponding to the room j, and C d is the total cost of the central air conditioner to be allocated in the current period.

[0046] As an optimization, the comfortable temperature can be taken as 25 degrees Celsius when cooling, and 18 degrees Celsius when heating.

[0047] ​​Preferably, the max() function corresponds to the room number which is obtained offline.

[0048] Preferably, in the d-th time period correction coefficient calculation, the input vector of the second neural network is The parameter values in the inner loop are the average values in the time period.

[0049] Preferably, the cooling capacity Q of the current time period is id It is obtained in a discrete short period accumulation manner.

[0050] Preferably, the input of the second neural network further includes a solar radiation intensity parameter.

[0051] Preferably, the central air conditioner is a water-cooled central air conditioner, the terminal unit is a terminal air damper, and the first neural network takes the central air conditioner host chilled water supply flow, supply and return water temperature difference, current temperature and humidity of the room, and a vector composed of the opening state values of all terminal air dampers as input.

[0052] Preferably, the central air conditioner is a full-air central air conditioner, the terminal unit is a terminal air damper, and the first neural network takes the central air conditioner host air supply flow, supply and return air temperature difference, current temperature and humidity of the room, and a vector composed of the opening state values of all terminal air dampers as input.

[0053] Preferably, the central air conditioner is a variable refrigerant flow central air conditioner, the terminal unit is an indoor unit, and the first neural network takes the central air conditioner host power, indoor unit power, current temperature and humidity of the room, and a vector composed of the opening state values of all terminal unit fans as input.

[0054] Preferably, when collecting the first data sample: set the target temperature according to the initial temperature of the room, control the terminal air damper fan driver to operate to adjust the room temperature to the target temperature and obtain the current humidity as the target humidity; after the room is kept at the target temperature for a period of time, the reference air conditioning unit and the terminal air damper are respectively used as a cold source to work for a certain time to collect samples, and the room temperature is kept at the target temperature when the two kinds of cold sources are working respectively, wherein the terminal unit fan works in PWM mode, and when the reference air conditioning unit works, the humidity is kept at the target humidity through the working condition adjusting unit,

[0055] Based on the working characteristics and working conditions of the reference air conditioning unit, the cooling capacity consumption power is calculated and divided by the average PWM duty cycle of the terminal air damper or the terminal unit fan, and the obtained value is taken as the unit time refrigeration capacity equivalent when the terminal air damper or the terminal unit fan is in the current opening state value F.

[0056] As preferred, when the room has multiple end units, the current room's current period cooling consumption is where F i is the set of all windbreaks in the current room.

[0057] As preferred, the cost allocation of the central air conditioner also includes an area allocation part, and the cost of the i-th room in the d-th period is:

[0058]

[0059] where C d0 is the basic cost of the central air conditioner.

[0060] As preferred, the control unit is provided with a main processing module, and the main processing module includes a sample extraction module for collecting data samples of the second neural network, the sample extraction module is configured to periodically and continuously collect the thermal load working condition and cooling consumption parameters of the central air conditioner cooling room, and search for a period in which the second neural network input quantity changes within a preset fluctuation threshold, and store the data of the selected period into the data sample set of the second neural network.

[0061] As preferred, the preset fluctuation threshold is ±5%.

[0062] As preferred, the preset fluctuation threshold includes a first fluctuation threshold and a second fluctuation threshold, and when the change of the second neural network input quantity in a period exceeds the first fluctuation threshold and is less than the second fluctuation threshold, the sample value of the input quantity in this period is taken as the time average value of the input quantity in this period.

[0063] As preferred, the outdoor temperature t w For example, the average outdoor temperature in the period T, i.e. its time average value, is As preferred, the first fluctuation threshold and the second fluctuation threshold are ±2% and ±5%, respectively.

[0064] As preferred, each sample data of the first neural network is collected as follows:

[0065] After the room is maintained at the target temperature for a period of time, the driver is turned off and the timing is started, the reference air conditioning unit is controlled to maintain the room temperature at the target temperature from t=0 to t=T1, and the working condition adjusting unit is controlled to maintain the room humidity at the target humidity, the refrigeration power q(t) is calculated based on the working characteristics of the reference air conditioning unit and the working condition, and the refrigeration amount of the first period is accumulated

[0066] The reference air conditioning unit is closed and the timer is reset. The fan damper opening state value F is set. The driver is controlled to work in PWM mode and the room temperature is maintained at the target temperature from t=0 to t=T2. The equivalent time is calculated where Δ(t) is the PWM value,

[0067] The driver is closed again and the timer is reset. The reference air conditioning unit is controlled to maintain the room temperature at the target temperature from t=0 to t=T1, and the working condition adjustment unit is controlled to maintain the room humidity at the target humidity. The third period equivalent cooling capacity is calculated again

[0068] The equivalent cooling capacity per unit time of the fan damper under the current working condition is calculated:

[0069] Preferably, the sensing detection unit is provided with a flow detection module at the inlet of the chilled water / cooling air supply main pipe of the central air conditioning host, a temperature detection module, and a temperature detection module connected to the outlet of the return water / return air main pipe of the host,

[0070] The multiple temperature detection modules at different positions in the room are arranged at the same height and on two vertical diagonal lines respectively.

[0071] Preferably, the number of temperature detection modules is 3-6, and the arrangement height is about 2 meters.

[0072] Preferably, the humidity is sensed by a humidity detection module arranged in the middle of the room return air pipe in the sensing detection unit,

[0073] Corresponding to the low, medium and high three gears of the fan speed, the fan damper opening state value can be respectively taken as three fan power normalized values corresponding to the three gears.

[0074] Preferably, when the driver works in PWM mode, the equivalent time can also be calculated by multiplying the normalized speed of the fan motor at different gear speeds by the integral of the on-duty ratio Δ(t) in the cycle Where k(t) is the normalized speed, and the value of other speeds is the ratio of the speed value to the highest speed.

[0075] Preferably, the temperature difference threshold is a value between 0.1℃ and 0.5℃.

[0076] Preferably, the temperature difference between the target temperature and the initial temperature is ≥5℃.

[0077] In the same central air conditioning system, for each type of terminal unit, a first neural network is established and training samples are collected.

[0078] As preferred, the reference air conditioning unit adopts a cold and hot air conditioner; if the ratio of the equivalent time dT to the length of the second time period T2 is less than the duty cycle threshold Δs, when collecting the sample, the reference air conditioning unit is also controlled to work in the heating mode from τ=0 to τ=T3 in the second time period, and the equivalent heat Q3 of heating is recorded Correspondingly, the equivalent cooling capacity per unit time of the damper under the current working condition is calculated:

[0079]

[0080] As preferred, an electric heating module can also be provided in the reference air conditioning unit, and the electric heating module is controlled to heat from τ=0 to τ=T3, and the equivalent heat Q3 of heating is recorded, where pr is the heating power of the electric heating module (kW or kJ / s), and PF is calculated similarly.

[0081] As preferred, in the T1 and T2 time ranges, the heating module can be turned on with known power and the heat is calculated.

[0082] As preferred, the rated cooling power of the reference air conditioning unit is 0.85-1.15 times the maximum cooling capacity of the terminal fan damper.

[0083] As preferred, for the same type of terminal unit, the horizontal and vertical distances from the terminal unit to the inlet of the main refrigerant supply pipe of the central air conditioning main unit are used for secondary subdivision into small categories, and for each small category, a first neural network is established and training samples are collected.

[0084] As preferred, the uniform temperature module in the working condition adjustment unit includes a base, a vertical rotating shaft, a bent arm, a horizontal rotating shaft, a pitchable bracket with two sections of movable connection arms connected by bolts, an extensible support rod connected between the outer ends of the two sections of arms at an acute angle with the axis of the horizontal rotating shaft, and a uniform temperature fan carried at the end of the pitchable bracket,

[0085] The control unit is also configured to control the operation of the uniform temperature module, so that the shaft of the uniform temperature fan moves in a spatial spiral line to deliver the cold air blown by the central air conditioning terminal unit and / or the reference air conditioning unit to all directions of the room, until the temperature difference of the multiple temperature detection modules in the room is less than the temperature difference threshold.

[0086] As preferred, the back of the uniform temperature fan has a fan cover.

[0087] As preferred, the sensing detection unit is provided with an image acquisition module;

[0088] The control unit further comprises an input module, an image processing module, a fan processing module, a mapping module and an output module, and the control unit is further configured to:

[0089] The image processing module analyzes the room orientation features based on the room image acquired by the image acquisition module, and extracts two mutually perpendicular diagonal lines;

[0090] The main processing module responds to events and schedules other modules;

[0091] The fan processing module adjusts the PWM wave duty cycle value of the driver according to the average of the plurality of temperatures at different positions in the room;

[0092] The mapping module has the first and second neural networks established therein, the input layers of the first and second neural networks respectively receive the input quantities from the main processing module, and the output quantities of the respective output layers are transmitted to the iterative learning part and the main processing module through the first and second connection arrays respectively; when the first and second neural networks are trained offline, the iterative learning part adjusts the connection weights of the first and second neural networks according to the actual value of the refrigerating capacity equivalent per unit time and the network output value input by the main processing module and the first and second neural networks through the first connection array respectively; when online metering, the first connection array is disconnected, the first and second neural networks respectively predict the refrigerating capacity equivalent per unit time and output to the main processing module through the second connection array, and the main processing module processes and analyzes the output and outputs through the output module.

[0093] As preferred, the control unit further comprises a working condition processing module, the working condition processing module comprises a uniform temperature planning part and a humidity adjusting part,

[0094] The uniform temperature planning part plans a spiral trajectory for the air direction of the uniform temperature fan in the uniform temperature module based on the diagonal lines and controls the uniform temperature module to blow air according to the trajectory at each time period,

[0095] Then, the moving speed of the uniform temperature fan according to the trajectory is controlled according to the temperature characteristics of the plurality of temperature detection modules in the room, and the moving speed is inversely proportional to the size of the temperature difference between the temperature in the corresponding direction and the target temperature.

[0096] As preferred, the reference air conditioning unit is provided with dry and wet bulb temperature detection modules and air supply amount detection modules, the sensible heat capacity under different working conditions is obtained according to standard tests, and the parameters are recorded as a working characteristic table or curve, based on the dry and wet bulb temperatures and the air supply amount of the current inlet and outlet air, the refrigerating capacity of the reference air conditioning unit in the current room is calculated by querying and interpolating the working characteristics.

[0097] As preferred, taking the water-cooled central air conditioner as an example, the first neural network adopts a BP neural network, and the model thereof is as follows:

[0098] The output of the jth node of the hidden layer is

[0099] The output of the output layer is

[0100] Wherein, x1-x4 are four scalars of the chilled water supply flow of the central air conditioner host, the supply and return water temperature difference, the current temperature and humidity of the room, and x5-xn are all the opening state values of the terminal units; the function f() is a sigmoid function, w ij and v j are the connection weights from the input layer to the hidden layer and the connection weights from the hidden layer to the output layer, respectively, θ j and θ are the threshold values of the hidden layer and the output layer, respectively, n and k are the node numbers of the input layer and the hidden layer, respectively, and the gradient descent method is used for network training.

[0101] In still another embodiment of the present application, a comfort-based central air conditioner billing system is also provided, which comprises:

[0102] a user interface unit for entering parameters, initiating operations and human-computer interaction;

[0103] a driver for driving the fan corresponding to the terminal unit of the central air conditioner;

[0104] a reference air conditioning unit for which the working characteristics have been pre-acquired and which is used as a reference for the cooling capacity;

[0105] a sensing detection unit for detecting the working conditions and environment of the central air conditioner and the reference air conditioning unit;

[0106] a working condition adjustment unit for adjusting the running working conditions of the terminal unit of the central air conditioner and the reference air conditioning unit;

[0107] a control unit connected with the driver, the reference air conditioning unit, the sensing detection unit, the user interface unit and the working condition adjustment unit; wherein the control unit is configured to:

[0108] first, the cooling equivalent of the terminal unit is modeled,

[0109] the room where the terminal unit is located is taken as the thermal load, the working condition of the central air conditioner system is taken as the input, and the cooling equivalent of the terminal unit of the room per unit time is taken as the output, a first neural network is established in the control unit,

[0110] The cooling supply amount of the terminal unit is equivalent to the cooling supply amount of the reference air conditioning unit in the system under the same heat load condition, the first data sample is collected under different working conditions, and the first neural network is trained;

[0111] Secondly, the central air conditioning cooling supply rooms are classified according to the orientation type, and a cooling consumption model is established for each type of orientation room,

[0112] The current room temperature and outdoor temperature are two scalar quantities, and the vector composed of the indoor temperature values of the current room in six directions of front, back, left, right, up and down is used as the input quantity, and the cooling consumption equivalent per unit time of the current room is used as the output quantity, and the second neural network is established in the control unit according to the room,

[0113] The terminal unit in the current room is adaptively controlled, and the room temperature is kept unchanged in stages, and when the working condition is approximately stable, the second data sample of the second neural network under different input conditions is collected, the output quantity in the sample is predicted by the first neural network, and the training is carried out based on the collected data sample;

[0114] Based on the offline established cooling supply and consumption models, the central air conditioner is charged by time period in online application,

[0115] First, according to the cooling consumption model, the correction coefficient of the current room working condition is calculated In the formula, i is the current room number, the room number i = 1 ~ N is any integer, N is the total number of rooms, p i , p i0 is the mapping output of the second neural network when is used as the input vector, is the current working condition of room i, The current room temperature and outdoor temperature are the same as , and the indoor temperature values of the current room in six directions are the current room temperature;

[0116] Then the cost of the i-th room in the d-th time period is calculated as:

[0117] Wherein, PF i (t) is the cumulative cooling supply amount of the current period, Q jd is the cooling supply amount of room j in the current period, k j3 is the corresponding coefficient of room j, C d is the total cost of the central air conditioner to be allocated in the current period.

[0118] As preferred, the driver is built in the end unit to drive the fan motor, and the control unit controls the end unit to work in PWM mode through the driver interface.

[0119] As preferred, the cost sharing calculation is replaced by:

[0120] First, the conversion coefficient of the current room condition is calculated Where i is the current room number, s i , s j are the areas of the i, j rooms, the room number i, j = 1 ~ N, N is the total number of rooms, p i , p j are the room i, j respectively with as the input vector , its respective mapping output of the second neural network, is the current condition of room i,

[0121] and calculate the comfort coefficient Where p ic is the room i with as the input vector , its mapping output of the second neural network, the condition The difference between and is that the current room temperature is a preset comfortable temperature such as 25 degrees Celsius;

[0122] Then calculate the cost of the i room in the d time period:

[0123] Where k j1 , k j2 is the coefficient corresponding to room j, C d is the total cost of the central air conditioner to be shared in this period.

[0124] Compared with the prior art, the device and system have the following advantages: the present application takes the rooms where the central air conditioner terminal units are distributed as the heat load, and takes the movable reference air conditioning unit which is calibrated in advance with high precision as the reference to calibrate the cold equivalent of the central air conditioner terminal unit; and the unit area cold consumption power of the terminal unit under the condition that the rooms in different directions are affected by the sun and adjacent rooms is measured, and the correction coefficient of each room is obtained accordingly, so that the billing purpose of the central air conditioner according to the actual cold consumption and comfort of the indoor basic heat load is realized. The present application takes the four scalars, i.e. the key working condition factors affecting the cooling of the central air conditioner terminal unit such as the refrigerating water supply flow of the host, the supply and return water temperature difference, the current temperature and humidity of the room, and the vector composed of the opening state values of all terminal units as the input, and takes the unit time cooling capacity of the current air damper of the room, i.e. the cooling equivalent as the output, to establish the first neural network as the terminal unit measurement mapping model; the model can reflect the influence of the actual working condition change of the central air conditioner on the cooling capacity of the terminal unit, and can dynamically reflect the cooling capacity change of the terminal unit, and overcomes the defect of the prior art that the fixed coefficient is used to estimate the cooling capacity of the air damper with actual change. At the same time, the measuring points of the supply water temperature difference and flow are only arranged at the inlet and outlet of the central air conditioner host refrigerating water supply main pipeline to replace the multi-point arrangement of each terminal, the detection of the large flow relative to the small flow of the terminal can reduce the relative error, and the significant reduction of the measuring points makes it possible to further reduce the measurement error by using high-precision temperature difference detection. In addition, the present application classifies the fan terminal according to the model and the horizontal and vertical distances from the inlet of the central air conditioner host refrigerating water supply main pipeline, and takes the opening state values of all terminal units as the input in the model, so that the decoupling of the mutual restriction between the terminal units of each air damper is realized. In the cost allocation, the room cooling model represented by the second neural network takes the air temperature and radiation factors as the input to compensate for the difference of the sun on the rooms in different directions, and the room temperature of the six-direction adjacent rooms as the input to reflect the influence of the adjacent room heat transfer on the cooling consumption, and finally the difference of the envelope structure is also eliminated through the correction coefficient.

[0125] The present application can accurately measure the actual cooling capacity of the central air conditioner terminal unit under different working conditions, and realizes the billing according to the effective cooling consumption of the indoor heat load, which is helpful for the energy saving and utilization of the air conditioner. BRIEF DESCRIPTION OF DRAWINGS

[0126] Figure 1A 、 Figure 1B The two structural schematic diagrams of the central air conditioner billing device and system based on the comfort of the present application;

[0127] Figure 2 The structural schematic diagram of the control unit of the present application;

[0128] Figure 3A The structural schematic diagram of the water-cooled central air conditioner,Figure 3B Structure diagram of variable refrigerant flow central air conditioner;

[0129] Figure 4A Structure diagram of reference air conditioning unit, Figure 4B Structure diagram of terminal fan unit;

[0130] Figure 5A 、 Figure 5B Distribution and temperature uniformity processing diagram of temperature detection module;

[0131] Figure 6 Structure diagram of temperature uniformity module;

[0132] Figure 7A Mapping principle diagram of cold quantity measurement of the application; Figure 7B Conversion principle diagram of room cooling;

[0133] Figure 8A Structure diagram of mapping module in the application, Figure 8B Structure diagram of first neural network;

[0134] Figure 9 Terminal unit temperature adjustment principle diagram.

[0135] In the figure: 1000 comfort-based central air conditioner billing system, 100 comfort-based central air conditioner billing device, 200 server, 300 terminal unit / terminal air damper, 400 driver, 500 chilled water pipe; 600 reference air conditioning unit;

[0136] 120 sensing detection unit, 130 working condition adjustment unit, 140 user interface unit, 150 control unit;

[0137] 121 temperature detection module, 122 flow detection module, 123 image acquisition module, 124 electric metering module;

[0138] 131 temperature uniformity module, 132 vertical rotating shaft, 133 bent arm, 134 horizontal rotating shaft, 135 telescopic support, 136 pitchable support, 137 fan cover, 138 temperature uniformity fan, 139 base;

[0139] 151 input module, 152 main processing module, 153 image processing module, 154 working condition processing module, 155 fan processing module, 156 output module, 157 storage module, 158 mapping module;

[0140] 1521 sample extraction unit, 1522 correction coefficient calculation unit, 1523 billing processing unit;

[0141] 1541 temperature uniformity processing unit, 1542 humidity adjustment unit;

[0142] 1581 first neural network, 1582 first connection matrix, 1583 iterative learning unit, 1584 second connection matrix;

[0143] 310 air fan, 320 fan coil, 330 return air inlet, 340 indoor unit;

[0144] 610 outdoor unit module, 620 indoor unit module. DETAILED DESCRIPTION

[0145] The preferred embodiments of the present application will be described in detail below with reference to the drawings, but the present application is not limited only to these embodiments. The present application encompasses any alternatives, modifications, equivalent methods and solutions made within the spirit and scope of the present application.

[0146] In order for the public to have a thorough understanding of the present application, specific details are described in the following preferred embodiments of the present application, and the present application can be fully understood without these details by those skilled in the art.

[0147] The present application is described in more detail below with reference to the accompanying drawings in the following paragraphs. It should be noted that the drawings are all in a simplified form and use non-precise proportions, only to facilitate, clearly assist the purpose of describing the embodiments of the present application.

[0148] Example 1:

[0149] A central air conditioner is an air conditioner that supplies refrigerant to different rooms through a main machine or unit, a water pipe, or a refrigerant pipe to achieve the purpose of indoor air conditioning. Referring to Figure 3A As shown, a water-cooled central air conditioner uses water as a refrigerant, and a full-air central air conditioner is a VAV variable air volume air conditioning system that controls and adjusts the temperature of an air conditioning area by changing the air supply volume in the air pipe rather than the air supply temperature, thereby adapting to changes in the load of the air conditioning area. Referring to Figure 3B As shown, a variable refrigerant flow central air conditioner is a VRV or VRF air conditioning system that controls the flow of refrigerant and achieves refrigeration or heating through direct evaporation or direct condensation of refrigerant. Compared with the two heat exchanges of full-air and water-cooled central air conditioners, the VRV or VRF central air conditioning system only needs one heat exchange, so it is more efficient, but its single-machine power is limited, so it is suitable for small-scale centralized cooling / warming occasions such as villas and local floors of office buildings.

[0150] Without loss of generality, the present embodiment first describes the apportionment billing of a water-cooled central air conditioning system. In a water-cooled central air conditioning system, the terminal device is a terminal unit, a terminal fan unit, or a terminal damper.

[0151] As shown in Fig. 1, the present application is based on the comfort of the central air conditioning billing device 100, which comprises: control unit 150, and with the control unit 150 connected to the sensing detection unit 120, user interface unit 140 and working condition adjustment unit 130. Among them, the present application with water-cooled central air conditioning terminal unit 300 in the room as the heat load, reference air conditioning unit 600 used as the reference for the room under different working conditions change cooling capacity, sensing detection unit 120 to the water-cooled central air conditioning and reference air conditioning unit 600 operating conditions, and the room cooling condition parameter detection, user interface unit 140 for typing parameters and initiate operation, including man-machine interaction when the display, such as the output of each household power display.

[0152] Water-cooled central air conditioning system consists of one or more cold heat source system and a plurality of terminal air conditioning system. Central air conditioning system running process is essentially a heat transfer process, reference Figure 3A 、 Figure 4B As shown, the cold / heat source is the host in which the host is cooled by the compressor, and the circulating water is cooled to chilled water after passing through the heat exchanger, and is delivered to each terminal air conditioning system, i.e. fan coil 320 in the figure; the chilled water is cooled to the room after passing through the fan coil, and the water temperature rises, and the circulating heat exchanger is cooled to the chilled water after being evaporated by the compressor refrigerant, thereby continuously removing heat from the room; at the same time, the compressor refrigerant is sucked into the compressor and compressed into high pressure steam, and then discharged to the condenser. The outdoor fan or cooling water system discharges the heat of the condenser, i.e. secondary heat exchange is carried out on the condenser, and the hot gas discharged from the condenser is discharged to the outdoor environment.

[0153] Reference Figure 4B Fan coil 320 is widely used in hotels, shopping malls, office buildings, hospitals, office buildings and other places, and is a working unit for primary heat exchange between indoor air and chilled water. Its working principle is that the fan makes indoor air or outdoor mixed air flow under the action of fan 310, flows through the surface cooler, i.e. the curved pipe through which the chilled water flows, is cooled and sent into the room, so that the indoor temperature is reduced to meet the comfort requirements of people; the cold air blown into the room is heated by the heat of the personnel, equipment and surrounding walls, and then passes through the return air port 330 and circulates to the fan coil 320 to be reheated.

[0154] In the heat transfer process of central air conditioning, the host delivers cold to each terminal unit. Reference Figure 7B As shown in Fig. 1, in order to charge each room for the cooling capacity consumed, two problems need to be solved: how much cooling capacity does each room consume?

[0155] However, how much difference is there in the comfort level of each room after consuming the same amount of cooling due to the influence of sunlight, enclosure structure, and adjacent rooms, and how to correct this difference? At the same time, different users set different target temperatures, and how to distinguish the billing of different temperatures?

[0156] Where does each room consume how much cooling through the end unit? This is the first question that must be answered for energy consumption billing.

[0157] Currently, the calculation of the refrigeration capacity of the end unit is based on the monitoring of the state of the three-speed switch of the fan, and the weighted sum of the working time of the high, medium and low speed gears is obtained, and the weight value, i.e. the coefficient, can only be obtained from the manufacturer's data under rated conditions.

[0158] The limitation of this fixed weight coefficient measurement method is obvious. First of all, due to the influence of installation conditions such as distance from the main machine, the actual air volume of different fans may differ from the nominal value of each gear; secondly, and more importantly, the water-cooled central air conditioning system including each end unit has a dynamic working condition, and using a fixed amount to calculate an actual changing amount is obviously unreasonable.

[0159] In the rooms using cooling, not only will there be a difference between the air gears and the nominal values, but also the total cooling capacity of the main machine will change, and the distribution of the total cooling capacity among the end units is not a simple linear proportional relationship but a mutual restraint among them, i.e. there is a nonlinear coupling relationship between the air gears.

[0160] Therefore, the present application regards the water-cooled central air conditioning system including each end unit as a whole, regards the distribution of the cooling capacity among the end units as a black box, and models the mapping relationship between the key working conditions of the system and the refrigeration equivalent of the end unit based on nonlinear modeling theory.

[0161] In order to identify the model, a data set for identification is needed. Among them, the cooling equivalent under different working conditions needs to be obtained. How to obtain the cooling equivalent? Currently, the enthalpy difference method of heat transfer medium is mostly used.

[0162] This method is first used in the central air conditioning billing system to construct a heat meter to measure the heat of heating. The heat meter consists of a hot water flow meter, a pair of temperature sensors and an integrator, and its working principle is that hot water provided by the heat source flows into the heat exchange system at a certain high temperature and flows out at a lower temperature, and in this process, heat is provided to the user. In a certain period of time, the heat obtained by the user can be calculated by the following equation:

[0163] E = ∫K (Ts-Tr) dV,

[0164] Wherein, E is the heat output of the heat exchange system, K is the correction coefficient of the specific heat and the specific gravity of the hot water, Ts and Tr are the supply water temperature and return water temperature respectively, and V is the hot water flow rate flowing through the heat supply system in a period of time.

[0165] The main error of the enthalpy difference method comes from the measurement of the working medium flow and the determination of the enthalpy value, especially the measurement error of small flow is large. Similarly, there is also a method of calculating the cooling capacity by detecting the supply and return air of the air side. Compared with the water side measurement, the air side measurement reduces the precision requirement of the temperature measurement equipment and instrument because the air supply temperature difference is much larger than the cold water supply and return water temperature difference, but no matter the air side or the water side measurement, at present, the measurement points are mainly set on the supply and return lines of the working medium at the end, and due to the small end flow and the fluctuation of the temperature and flow parameters being much larger than that of the main machine end, there is a contradiction between the sensor precision and the instrument equipment cost.

[0166] Based on the above research, in order to improve the model generalization ability and prediction accuracy, the present application takes the chilled water supply flow of the water-cooled central air conditioner main machine, the supply and return water temperature difference, the current temperature and humidity of the room, and the vector composed of the opening state values of all terminal units as input quantities, and takes the equivalent cooling capacity of the current air damper of the room per unit time as the output quantity, and establishes a first neural network in the control unit as a terminal unit measurement mapping model. Among them, the two key influencing factors of flow and temperature difference only need one measurement point, and more importantly, the detected is the flow of the main pipeline, which is much larger than the end flow, which can effectively improve the measurement accuracy.

[0167] The present application takes the room where the terminal unit of the water-cooled central air conditioner is located as the heat load, and takes the reference air conditioning unit as the reference object of the changing cooling capacity of the room under different working conditions to obtain the cooling capacity equivalent value in the data sample required for system identification.

[0168] Specifically, as shown in Figure 7A , the reference air conditioning unit, i.e. the reference machine, selects an integrated movable air conditioner, first obtains the sensible cooling capacity under different working conditions according to the standard test, and records the parameters as a working characteristic table or curve, establishes a second mapping of the working condition of the reference machine to the cooling capacity, and provides a basis for sample cooling capacity calculation in the working room at the end of the air damper.

[0169] Then, under the condition of each input combination, the training sample of the established first neural network is obtained offline. According to the characteristics of the cooling capacity transmission from the main machine to the terminal unit of the water-cooled central air conditioning cooling system, in order to solve the influence of system time lag and large inertia, combined with Figure 4A , Figure 4B As shown in

[0170] The control unit drives the fan corresponding to the water-cooled central air conditioner terminal unit through the driver;

[0171] Taking the room where the water-cooled central air conditioner terminal unit is located as the thermal load, a target temperature is set according to the initial temperature of the room; the temperature difference between the multiple temperature detection modules in the sensing detection unit located at different positions in the room is less than the temperature difference threshold through the working condition adjustment unit; the driver is controlled to adjust the room temperature to the target temperature by the central air conditioner terminal fan and obtain the current humidity as the target humidity,

[0172] After the room is kept at the target temperature for a period of time, the driver is turned off and timing is started, the reference air conditioning unit is controlled to maintain the room temperature at the target temperature from t=0 to t=T1, and the working condition adjustment unit is controlled to keep the room humidity at the target humidity, the refrigeration power q(t) is calculated based on the working characteristics and working conditions of the reference air conditioning unit, and the refrigeration amount in the first period is accumulated

[0173] The reference air conditioning unit is turned off and timing is started again, the opening state value F of the fan damper is set, the driver is controlled to work in PWM mode and maintain the room temperature at the target temperature from t=0 to t=T2, and the equivalent time is calculated Where Δ(t) is the PWM value,

[0174] The driver is turned off again and timing is started again, the reference air conditioning unit is controlled to maintain the room temperature at the target temperature from t=0 to t=T1, and the working condition adjustment unit is controlled to keep the room humidity at the target humidity, and the refrigeration amount in the third period is calculated again

[0175] The equivalent refrigeration amount per unit time of the damper under the current working condition is calculated:

[0176] As a preferred, the length of the third period can be different from the length of the first period, such as the difference between the two is within 20%. As a preferred, the opening state value F is three gears of low speed, medium speed and high speed respectively.

[0177] During the continuous sampling of the sample set, only the working condition of the water-cooled central air conditioning system can be changed and the data in the second period can be collected, and the first and third periods can be collected again every few samples. The first sample collection after changing the target temperature is performed in three periods.

[0178] Combined Figure 7AAs shown in the figures, the present application establishes a first neural network in the control unit as a first mapping of the end unit working condition state of the central air conditioning system to the refrigeration capacity / cold equivalent, in order to avoid errors caused by small flow of the end, and the measured point parameters at the large flow of the main pipeline are taken as the input quantity of the mapping; and in the working room, the end unit and the reference air conditioning unit are alternately refrigerated, and the end unit to be metered is calculated based on equivalent refrigeration, that is, the cold obtained by the reference air conditioning unit through the second mapping under the same working condition is taken as the cold equivalent of the end unit. During this period, the heat load conditions of the two cold sources, that is, the room inflow heat flux density under cooling, are made the same through sensing and control of the working condition, so as to ensure the reliability of the equivalent calculation.

[0179] Without loss of generality, when the water-cooled central air conditioning end unit is stably working online, the humidity of the room will basically remain stable after the main machine is set, mainly affected by the seasonal climate. Therefore, the current humidity of the room during cooling of the central air conditioner is taken as the target humidity, and the humidity of the room is maintained at the target humidity during cooling of the reference air conditioning unit.

[0180] For the room temperature, the target temperature needs to be set according to the initial temperature of the room under natural conditions without cooling. As preferred, the temperature difference between the target temperature and the initial temperature is ≥5℃, and the load rate of the end unit during sample collection is greater than a set value, such as the power reaching 0.5-1 times of the rated power.

[0181] During cooling, the temperature gradient will appear in the room due to the proportion of the cold air inlet to the room surface area, which may cause deviation of the heat load during working of the two cold sources. In order to reduce the deviation of the heat load, refer to Figure 5A 、 Figure 5B , and combine Figure 6 As shown in the figures, a plurality of temperature detection modules 121 are arranged at different positions in the room, and the temperature difference of these temperature detection modules is less than a temperature difference threshold value through the uniform temperature module 131 in the working condition adjustment unit.

[0182] Specifically, as shown in Figure 6 , the working condition adjustment unit is provided with a uniform temperature module 131, which includes a base 139, a vertical rotating shaft 132, a bent arm 133, a horizontal rotating shaft 134, a pitchable support 136 with two sections of support arms connected by bolts, and a telescopic support rod 135 connected between the outer ends of the two sections of support arms and forming an acute angle with the axis of the horizontal rotating shaft 134, and a uniform temperature fan 138 is carried at the end of the pitchable support 136. As preferred, the back of the uniform temperature fan 138 is provided with a fan cover 137. An image acquisition module 123 is arranged in the sensing and detection unit, which can be arranged at the bottom of the bent arm 133, so as to obtain the global image of the room through rotation of the vertical rotating shaft 132.

[0183] Referring to Figure 2 As shown in the figure, preferably, the control unit 150 comprises an input module 151, a main processing module 152, an image processing module 153, a working condition processing module 154, a fan processing module 155, a mapping module 158 and an output module 157, wherein the working condition processing module 154 further comprises a uniform temperature planning part 1541 and a humidity adjusting part 1542. The control unit is further configured to:

[0184] The main processing module responds to events and schedules other modules;

[0185] Based on the room image obtained by the image acquisition module, the image processing module 153 analyzes the room orientation features and extracts two mutually perpendicular diagonal lines. The orientation features include the distribution and length of the room structure edge lines, as well as the directions and distances of the cold source air outlet, the return air inlet and the uniform temperature module 131 in the room relative to the corners of the room.

[0186] In combination with Figure 5A , Figure 5B As shown in the figure, the uniform temperature planning part 1541 in the working condition processing module 154 plans the operation trajectory for the uniform temperature module 131 based on the above-mentioned orientation features, so that the axis of the uniform temperature fan 138, i.e. the end of its pointing direction, moves in a spatial spiral line to deliver the cold air blown out by the central air conditioning terminal unit and / or the reference air conditioning unit to each area of the room, until the temperature difference of the multiple temperature detection modules in the room is less than the temperature difference threshold.

[0187] The planning of the trajectory can be based on the obtained room diagonal lines, taking the line connecting the cold air outlet position to the farthest place in the room that it can reach or its opposite side as one of the main diagonal lines, and taking a straight line perpendicular to it as the other diagonal line; then, planning a spiral trajectory with the main diagonal line as the axis. In Figure 5A , the black dot is the air outlet located in the corner, and a curve spirally moving around the inner wall surface of the cone is planned as the target trajectory with the diagonal line where the air outlet is located as the center, wherein the main diagonal line is the center line of the cone. In Figure 5B , the air outlet is located in the middle of one side wall, and a curve spirally moving around the inner wall surface of the cylinder is planned as the target trajectory with the side wall as the ground of the cylinder, wherein the main diagonal line is the center line of the cylinder. The planned target trajectory should avoid the return air inlet to avoid cold energy loss.

[0188] As a preferred, typical indoor orientation features can be summarized and classified, the end trajectory curve is planned for each orientation category based on geometric equations, and the output angles of each joint in the uniform temperature module including the vertical rotation axis, the horizontal rotation axis and the telescopic strut are analyzed based on inverse kinematics.

[0189] As preferred, the joint angle data sequence corresponding to the trajectory can also be stored by on-site demonstration, and the joint control is performed according to the sequence.

[0190] The movement according to the planned trajectory can make the temperature of each area of the room be reduced as soon as possible, and reduce the temperature gradient between the areas. The uniform temperature planning unit controls the uniform temperature module according to the planned trajectory, and in each time period of sample collection, firstly, the uniform temperature fan is operated at a uniform speed to reduce the temperature generally; in order to further reduce the temperature difference between the areas, as preferred, then, according to the temperature characteristics of the multiple temperature detection modules in the room, the linear speed of the uniform temperature fan moving according to the trajectory is changed, and the linear speed is inversely proportional to the size of the temperature difference between the temperature at the trajectory point and the target temperature. The temperature at each trajectory point can be obtained by interpolation calculation according to the temperature values of the multiple temperature detection points. Through the speed planning, the uniformity of each space point of the room temperature can be improved, so that the working condition consistency is ensured, and the generalization ability of the measurement model is improved.

[0191] The uniform temperature module is periodically operated, and when the temperature difference between the highest temperature and the lowest temperature of the multiple temperature detection modules is less than a temperature difference threshold, the operation is stopped and the data sample can be collected; when the temperature difference is monitored to be greater than the threshold, the operation is started again so that the overall temperature of the room is dynamically balanced at the target temperature. As preferred, the temperature difference threshold is a value between 0.1℃ and 0.3℃.

[0192] In a second time period during sample collection, the fan processing module in the control unit adjusts the PWM duty cycle value of the driver according to the average of the multiple temperatures at different positions in the room, that is, the overall temperature of the room. Figure 9 As shown, according to the error value e(t) between the target temperature and the current overall temperature of the room, the fan processing module in the host unit calculates the PWM value connected to the fan driver based on the PID control law, changes the driving power width of the driver to change the operation speed of the fan, so that the error value e(t) dynamically approaches 0.

[0193] The working condition processing module also has a humidity adjusting unit, which controls the operation of the humidity adjusting module in the working condition adjusting unit based on the monitoring of the humidity detection point in the room, so that the humidity of the room is kept at the target humidity.

[0194] In combination with Figure 3A , Figure 5A , Figure 5B As shown, the humidity detection module is arranged in the middle of the return air duct of the room; the flow detection module 122 and the temperature detection module 123 of the water are arranged at the inlet of the main pipe of the chilled water supply of the central air conditioner host; the temperature detection module of the water is arranged at the outlet end of the return water main pipe connected to the host; multiple temperature detection modules are arranged at different positions in the room, which can be arranged at the same height and respectively located on two perpendicular diagonal lines.

[0195] As preferred, the plurality of temperature detection modules is 3-6, and the setting height is about 2 meters; the temperature values of the plurality of temperature detection modules can be averaged to obtain the current temperature of the room.

[0196] During sample collection, the central air conditioner host can be operated to adjust the power to keep the chilled water supply flow and the supply and return water temperature difference constant. As preferred, the two parameters can be taken as the average values during the sample collection period based on the proportional relationship between the cooling capacity and the temperature difference and flow.

[0197] As preferred, the plurality of temperature detection modules set for controlling the consistency of the heat loads of the two cold sources are only used for sample collection, so one of the temperature detection modules, such as the module near the return air inlet temperature measurement point, can be selected as the current temperature of the room in the first neural network input quantity in online application, thereby simplifying the system structure and facilitating actual operation.

[0198] For each terminal unit of the water-cooled central air conditioner, the wind deflector opening state value can be taken as the normalized value of the fan power corresponding to the low, medium and high three gears of the fan speed, for example, the highest power value is taken as 1, and the power of the other two gears is scaled proportionally.

[0199] The cooling system of the water-cooled central air conditioner is a nonlinear lag system, so changes in operating parameters need a period of time to feedback their effects. Therefore, the present application sets sampling conditions when collecting training samples, allowing the system to enter a steady state before sampling, and allowing the room conditions to remain for a period of time when the reference air conditioning unit and the terminal unit are cooled to eliminate the randomness of short-term sampling. At the same time, by sampling the reference air conditioning unit before and after the terminal unit is cooled, the influence of slow fluctuations in working conditions on the sampling data is eliminated, and the prediction accuracy of the network model is improved.

[0200] The first neural network established by training the collected sample set predicts the equivalent cooling capacity of the current wind deflector per unit time in the field environment using the trained first neural network, and outputs the predicted value through the output module. This value can be used as the basis for charging the water-cooled central air conditioner for each terminal.

[0201] As shown in Figure 7A , the working room is the heat load, and the equivalent cooling capacity provided by the terminal unit under the same working condition is obtained through the second mapping of the reference air conditioning unit. Therefore, the working characteristics of the movable reference air conditioning unit need to be obtained through high-precision calibration in advance.

[0202] Referring to Figure 4AAs shown, the reference air conditioning unit 600 comprises an outdoor unit module 610 and an indoor unit module 620. In the reference air conditioning unit, a dry-bulb temperature detection module and a supply air volume detection module can be provided; a test device for establishing a room air enthalpy method is established according to a room air conditioner standard, and sensible heat release thereof under different working conditions is tested and recorded as a working characteristic table or curve as the working characteristic of the reference air conditioning unit. Then, when a sample is collected, based on the dry-bulb temperature and the wet-bulb temperature of the current inlet and outlet air and the supply air volume, the refrigerating capacity of the reference air conditioning unit under the current working condition is calculated through querying and interpolating the working characteristic.

[0203] In the air enthalpy method, the calculation formula of the sensible heat release is:

[0204] φ sc =q m ·c pa ·(t a1 -t a2 ) / V n ·(1+W n ),

[0205] wherein φ sc is the sensible heat release (W), q m is the supply air volume (m 3 / s) of the measuring point, V n is the specific volume of the wet air (m 3 / kg) at the measuring point, W n is the air humidity at the measuring point, t a1 and t a2 are the return air temperature and the supply air temperature (℃) respectively, and the constant-pressure specific heat capacity c pa =1005+1846W n (J / (kg·K)).

[0206] In the reference air conditioning unit, the dry-bulb temperature and the wet-bulb temperature are detected by sensors placed in the supply air inlet and the return air outlet in a heat preservation section. The dry-bulb and the wet-bulb are used to detect the supply air temperature, the return air temperature and the humidity, and these parameters can also be obtained by using a temperature sensor and a relative humidity sensor. During the collection of the training samples, the reference air conditioning unit adjusts the operating frequency under the instruction of the control unit to maintain the temperature of the room at the target temperature.

[0207] As a preferred, the working characteristic data of the reference air conditioning unit is recorded in the form of a table, and the sensible heat release under the current working condition is calculated based on the working characteristic table lookup and multi-dimensional interpolation.

[0208] As a preferred, the rated refrigerating capacity of the reference air conditioning unit is 0.85-1.15 times the maximum air volume refrigerating capacity of the terminal fan.

[0209] In combinationFigure 8A 、 Figure 8B As shown in FIG. 16, the control unit establishes a first neural network 1581 as an end unit metering mapping model in the mapping module 158, the input layer of the first neural network receives input quantities from the main processing module 152, and the output quantities of the output layer are transmitted to the iterative learning part 1583 and the main processing module 152 through the first connection matrix 1582 and the second connection matrix 1584, respectively; when training the first neural network offline, the iterative learning part 1583 adjusts the connection weights of the first neural network 1581 until the learning is completed according to the wind deflector unit time refrigeration quantity equivalent actual value and the network output value input by the main processing module 152 and the first neural network 1581 through the first connection matrix 1582, respectively; when metering online, the first connection matrix 1582 is disconnected, the first neural network 1581 predicts the wind deflector unit time refrigeration quantity equivalent and outputs it to the main processing module 152 through the second connection matrix 1584, and the main processing module 152 processes and analyzes it and outputs it through the output module 156. As shown in FIG. 17, Figure 1A 、 Figure 2 As shown in FIG. 17, the output module can transmit the metering result to the user interface unit 140 for display or store it in the server 200, wherein the server 200 can realize communication with one or more systems of the application through a cloud platform. In the main processing module 152, the sample extraction part 1521 controls the collection and screening of training samples, the correction coefficient calculation part 1522 calculates various coefficients for apportioning charges, and the charge processing part 1523 apportions the central air conditioning charges based on the working conditions of each end unit in a distributed processing manner.

[0210] As a preferred, the first neural network adopts a BP first neural network, and the model thereof is as follows:

[0211] The output of the jth node of the hidden layer is

[0212] The output of the output layer is

[0213] Wherein, x1-x4 are four scalars of the central air conditioning host refrigerated water supply flow, supply and return water temperature difference, current temperature and humidity of the room, x5-xn are all end unit opening state values; the function f() is a sigmoid function, w ij and v j are the connection weights from the input layer to the hidden layer and the connection weights from the hidden layer to the output layer, respectively, θ j and θ are the threshold values of the hidden layer and the output layer, respectively, n and k are the node numbers of the input layer and the hidden layer, respectively, and the gradient descent method is used for network training.

[0214] As a preferred, other redundant factors such as the end unit supply air temperature can also be added to the network input quantities.

[0215] Because the chilled water supply pipe is shared, the water supply pressure will decrease step by step with the distribution of the chilled water. The actual cooling capacity of the water-cooled central air conditioning fan coil end unit is not only related to the model and the air damper position, but also related to the distance from the main unit. Therefore, for the same model end unit, the horizontal distance and the vertical distance from the end air damper to the chilled water supply main pipe inlet of the central air conditioning main unit can be used to further subdivide the end unit into small categories. A first neural network is established for each small category, and the training samples are collected respectively, so that the prediction of the cooling capacity equivalent is more accurate.

[0216] As preferred, for each category, one end unit is selected for sample acquisition and training. For the first neural network of each category end unit, a room with slow and small peripheral temperature change is selected as the heat load during the collection of training samples, so that the sample collection can be continuously performed, and the sampling time of the overall sample set is shortened. For this purpose, the rooms inside the building can be used as the collection environment. As preferred, for the rooms located at the corners of the building, the sample collection is performed at night or during the time without direct sunlight during the day.

[0217] Cooling capacity measurement is the basis for charging the central air conditioning system according to the cooling capacity. However, in addition to the measurement, many factors need to be considered for charging. For example, Figure 7B As shown in the figure, the cooling capacity required for cooling per unit area is very different for rooms in different positions in the building. For example, the first floor lobby has less envelope structure, and more cold air is delivered to the environment. The rooms at the edges and the top are obviously more affected by solar radiation than the rooms in the middle, and therefore require more cooling capacity. Obviously, if the actual cooling capacity consumption is simply used for charging, it is unfair to the rooms that consume more cooling capacity due to environmental factors.

[0218] Therefore, the cooling characteristics of rooms in different positions in the building must be analyzed first, and the cooling capacity of rooms in different positions must be compensated and converted according to the cooling characteristics. First of all, how to describe the cooling characteristics of rooms in different positions needs to be studied, such as whether a model can be established and how to establish the model. Such a model should reflect the main and quantifiable factors that cause the change in cooling capacity consumption in different positions, and these factors should be able to be converted to correct or compensate different rooms. Then, even if a model can be established, how to correct it based on the model? After a lot of analysis and research, we found that such correction should reflect the influence of environmental factors, and the current temperature set value or actual value of each room should be excluded from the body of such correction, otherwise it will return to the charging mode of using more or less the same.

[0219] For this purpose, see Figure 7BAs shown, this invention uses two scalar quantities—the current room temperature and the outdoor temperature—as well as a vector composed of the room's indoor temperature values ​​in six directions (front, back, left, right, up, and down) during cooling as input quantities. The equivalent cooling capacity per unit time of the room's fan speed is used as the output quantity. A second neural network is established in the control unit for each room, serving as a cooling consumption model for each room. Then, based on this model, the equivalent cooling capacity per unit time for each room under near-optimal operating conditions is predicted, and a correction coefficient is calculated accordingly. Finally, based on this correction coefficient, the actual cooling capacity predicted by the first neural network is corrected, and the deferred costs of the entire water-cooled central air conditioning system are allocated accordingly.

[0220] Specifically, refer to Figure 8A , Figure 8B The modeling process of the first neural network is described, and the second neural network can also use a BP network. Of its nine inputs, x1-x2 represent the current room temperature, outdoor air temperature, and solar radiation intensity (three scalars); x3-x8 represent the room temperature in six directions (front, back, left, right, top, bottom, and front / back). The output y(t) represents the equivalent cooling load per unit time in the room. To enable the prediction and calculation of this equivalent cooling load (i.e., the required cooling load) using the cooling equivalent of the first neural network, this invention maintains a dynamically stable room temperature during sample collection to achieve a balance between cooling supply and demand. Therefore, the output of the second neural network is the equivalent cooling load per unit time of the room's fan speed. When the terminal unit provides cooling, room humidity is mainly affected by seasonal climate; however, preferably, the second neural network can also include indoor humidity as an input. When adjacent rooms are empty, the temperature in that direction, such as the outdoor air temperature, is used as the indoor temperature of that room. Preferably, when the water-cooled central air conditioning system enters a stable operating state, the six-directional room temperature values ​​in the second neural network can be replaced with the on / off status values ​​of the terminal unit fans in the six-directional rooms. Preferably, a solar radiation intensity parameter is also added to the input of the second neural network.

[0221] When the operating conditions are approximately stable, data samples of the second neural network under different input conditions are collected. The output is predicted by the first neural network corresponding to the end unit. The second neural network is trained based on the data samples and is used to predict the equivalent cooling load per unit time of each room under the current operating conditions.

[0222] In order to realize the automatic collection of the second neural network data sample, the application is provided with a sample extraction module in the main processing module of the control unit, which is used to collect the thermal load working condition and the cooling consumption parameter of the water-cooled central air conditioning cooling room periodically and continuously, and search for the period in which the variation range of each input variable of the second neural network is within the preset fluctuation threshold, and store the data of the selected period into the data sample set of the second neural network. Through the automatic extraction of the sample, the sample can be obtained during the working process of the water-cooled central air conditioning, and it is not necessary to specially start up for a long time for debugging, so that the data collection cost is greatly saved.

[0223] As preferred, the preset fluctuation threshold is ±5%, and the value of each input sample parameter is the arithmetic mean value or the median value in the period.

[0224] As preferred, the preset fluctuation threshold can also include two sets of threshold ranges of the first fluctuation threshold and the second fluctuation threshold, and when the variation of the second neural network input variable in the period exceeds the first fluctuation threshold and is less than the second fluctuation threshold, the sample value of the input variable in the period is taken as the time average value of the input variable in the period. For example, the average outdoor temperature t w in the period T is the time average value of

[0225] As preferred, the first fluctuation threshold and the second fluctuation threshold are ±2% and ±5% respectively.

[0226] When the second neural network is established, a model can be established for each room respectively; as preferred, the same network can be used to represent the same type of rooms in the same orientation feature, such as the middle rooms. Through the sharing of the model, the sample amount can be further reduced, and the calculation and modeling time can be saved.

[0227] After the above two modeling is completed offline, the water-cooled central air conditioning is cost allocated in different time periods in online application. Different from the existing method, when the cooling consumption of the rooms in different orientations is corrected, the application does not use the fixed, statistical empirical value as the coefficient, but real-time conversion is carried out according to the actual working condition of each room.

[0228] Specifically, the correction coefficient of the current room working condition is calculated according to the cooling consumption model In the formula, i is the current room number, the room number i is any integer in 1 to N, N is the total number of rooms, p i , p i0 is the mapping output of the second neural network when is taken as the input vector, is the current working condition of the room i, the current temperature of the room and the outdoor temperature and The room indoor temperature value in the six directions of the same and current room is the current room temperature;

[0229] The i room cost in the d time period is calculated again as:

[0230] Wherein, PF is the predicted end unit cooling capacity equivalent per unit time in the current room based on the first neural network i Q is the accumulated current period cooling capacity of the current room jd Q is the current period cooling capacity of room j j3 C is the corresponding coefficient of room j d C is the total cost to be allocated of the central air conditioner in the current period.

[0231] The correction coefficient is necessarily related to the cooling consumption equivalent of other rooms and the room temperature of other rooms at the same time due to the differences in the starting time, environment and set temperature. Therefore, when the correction coefficient of a room is calculated, is the cooling consumption equivalent calculated based on the working condition of the room?

[0232] Through in-depth research, the application uses the generalization characteristics of the room cooling model to uniformly predict the cooling consumption of the same room in the working condition without additional heat load of adjacent rooms. The predicted value is actually the cooling demand of the internal basic heat load of the room, and the ratio of the cooling demand corresponding to the internal basic heat load of the room to the actual cooling demand is taken as the correction coefficient k i3 ; in the central air conditioner cost allocation calculation formula, the actual cooling capacity of the room is multiplied by the correction coefficient k i3 , so as to realize the allocation of the cooling demand corresponding to the internal basic heat load of each room in the billing, and reflect the fairness of the billing. It can be understood that according to the cost allocation calculation formula, the correction coefficient k i3 of the room near the edge of the building, the top and bottom floors and the like will be smaller than that of the central room of the building, so that the coefficient can be used to obtain billing compensation.

[0233] At the same time, in different periods, if the set temperature of the same room is low, the actual cooling capacity will be large, so that more costs will be allocated. Therefore, the device also reflects the billing according to the comfort degree, which is helpful to guide the reasonable consumption of cooling and achieve the energy saving effect.

[0234] As a preferred, in the correction coefficient calculation of the d time period, the input vector of the second neural network is the average value in the period; and the current period cooling capacity Q id can be obtained in the form of short period accumulation in a discrete period system. As a preferred, The outdoor temperature can be taken as the current room temperature. The total cost to be allocated can be calculated by the consumption of electricity, water, etc.

[0235] When there are multiple terminal units in the room, the current room cooling consumption in the period where F i is the set of all air dampers in the current room.

[0236] It can be understood that, compared with water-cooled central air conditioning, all-air central air conditioning uses air instead of cooling water as refrigerant. Therefore, the water-cooled central air conditioning billing device scheme based on comfort in the embodiment is also applicable to all-air central air conditioning. Due to the difference in refrigeration process details, in the control unit of the billing device, the first neural network is modified to use the supply air flow rate of the all-air central air conditioning host air outlet, the supply and return air temperature difference, the current temperature and humidity of the room, and the vector composed of the opening state values of all terminal air dampers as input quantities.

[0237] Embodiment 2:

[0238] In the process of collecting the first neural network training samples, if the target temperature is initially set too high or too low, or the power of the terminal unit does not match the heat load of the room, it will cause the terminal fan working condition to be limited to low power or high power state when collecting samples.

[0239] Therefore, in order to make the samples cover different load rates of the terminal fan from low to high, in the embodiment, when the control unit controls the driver to work in PWM mode, during the collection of a sample, the terminal unit is switched to work in different gears, and the equivalent time of the terminal fan is calculated by multiplying the normalized speed of the fan motor at different gear wind speeds by the integral of the on-duty ratio Δ(t) in the period where k(t) is the normalized speed, and when the highest speed is 1, the value of other speeds is the ratio of the speed value to the highest speed.

[0240] Based on the same reason, the embodiment can also use the following method to obtain the training samples. The reference air conditioning unit uses a cooling and heating air conditioner; if the ratio of the equivalent time dT to the length of the second period T2 is less than the duty cycle threshold Δs, during the collection of the sample, the reference air conditioning unit is also controlled to work in heating mode from τ=0 to τ=T3 in the second period, and the equivalent heat of heating is recorded Correspondingly, the equivalent refrigeration capacity per unit time of the air damper in the current working condition is calculated as follows:

[0241]

[0242] As preferred, an electric heating module can also be provided in the reference air conditioning unit, and the electric heating module is controlled to heat from t=0 to t=T3, and the heat equivalent Q3=pr·T3 is recorded, where pr is the heating power of the electric heating module (kW or kJ / s), and PF is calculated similarly.

[0243] As preferred, in the time range of T1, T2, the heating module can be turned on with known power and heat calculation is performed, so as to expand the working condition coverage of the reference air conditioning unit and the water-cooled central air conditioning terminal unit.

[0244] The working characteristics of the reference air conditioning unit is the mapping between the working condition and the sensible cooling capacity. In this embodiment, the mapping can also be represented by a third neural network, and the input of the network can be selected as the dry and wet bulb temperatures of the inlet and outlet air and the supply air volume; or can be preferably the electric power, the outdoor condenser temperature, the indoor evaporator temperature and humidity; or can be preferably the compressor operating frequency, the supply air fan power, the supply and return air temperature and humidity.

[0245] Considering the central air conditioning equipment and installation costs, and the fact that there are some public rooms in the building, etc., the existing charges are composed of fixed charges and operating charges. The costs of the air conditioning system mainly include the following aspects: a. The electricity cost of the air conditioning system operation (including the electricity cost of the chiller, cooling tower, water pump and other equipment); b. The consumption of water supplement of the air conditioning system; c. The operation and management cost of the air conditioning system; d. The depreciation cost of the air conditioning system; e. The maintenance and repair cost of the air conditioning system; f. Other additional costs and property management costs. Among the above types of costs, the first three are the operating costs of the central air conditioning system, and the last three are basically a fixed value during the operation of the air conditioning system, which is called basic fee.

[0246] Different from embodiment 1, according to the combined charging method, the total cost to be allocated of the water-cooled central air conditioning is divided into two parts, in addition to the actual cooling capacity allocation, there is also an area allocation part, then the cost of the i room in the d time period is:

[0247]

[0248] Wherein, C d1 and C d0 are the operating cost and the basic cost of the central air conditioning respectively.

[0249] Embodiment 3:

[0250] This embodiment provides another allocation and charging calculation method, and takes a variable refrigerant flow multi-connected air conditioner as the central air conditioner as an example.

[0251] Variable refrigerant flow central air conditioning, also known as multi-connected central air conditioning system, is composed of a single refrigeration / heat cycle air conditioning system, also known as VRV or VRF, which is connected to a plurality of different or same type and capacity direct expansion indoor units from a single outdoor unit.

[0252] Referring to Figure 3B As shown, VRF is intended to be directly expanded, which directly from the host to the indoor unit without rough air pipe, water pipe, but with thin copper pipe to transport low-temperature liquid refrigerant after compression and expansion in the outdoor unit to the indoor unit as the end unit, and then into the evaporator through the electronic expansion valve to become gas and take away heat at the same time.

[0253] In the VRF system, the refrigerant flow is variable, which is determined by the indoor load. When the indoor load is high, a large amount of refrigerant is transported to the indoor unit to provide more cooling / heat; otherwise. The compressor of the host adjusts the refrigerant flow required by the whole system, and the refrigerant flow of the individual indoor unit is adjusted by the electronic expansion valve. In the indoor unit, the indoor air supply volume is also adjusted by the fan.

[0254] Different from example 1, referring to Figure 1B As shown, in the VRF system, the diameter of the refrigerant conveying pipe is much smaller than that of the water pipe and the air pipe, and the flow is smaller. Therefore, an electric metering module 124 is connected in the power supply circuit of each indoor unit 340 as the end unit.

[0255] Correspondingly, when metering and modeling the cooling equivalent of the end unit, the first neural network takes the central air conditioner host power, indoor unit power, current temperature and humidity of the room, and a vector composed of the on-off state values of all end unit fans as input, and takes the cooling equivalent of the end unit per unit time, i.e. the refrigeration equivalent, as output.

[0256] After completing the two modeling of the first and second neural networks for the end unit cooling and room cooling, in the online application, the embodiment adopts another allocation method to charge the water-cooled central air conditioner by time period.

[0257] Specifically, the conversion coefficient of the current room working condition is calculated In the formula, i is the current room number, s i , s j are the areas of the i-th and j-th rooms, respectively, the room number i, j=1~N, N is the total number of rooms, p i , p j are the conversion coefficients of the i-th and j-th rooms, respectively, when the input vector is the respective mapping output of the second neural network, for the current working condition of room i,

[0258] and calculate the comfort coefficient wherein p ic is the total cooling load of room i in the current working condition as the input vector and its mapping output corresponding to the second neural network, the working condition is the same as except that the current indoor temperature of the room is a preset comfortable temperature, such as 25 degrees Celsius;

[0259] The cost of the i-th room in the d-th time period is calculated as follows:

[0260] wherein k j1 , k j2 is the corresponding coefficient of room j, and C d is the total cost of the central air conditioner to be allocated in the current period.

[0261] In calculating the conversion coefficient k i1 , the room with the maximum cooling consumption per unit area is found by the max() function. Since the input quantity when obtaining the cooling consumption by mapping is the same, i.e., the working condition is the same, the denominator corresponds to the room with the same comfort level, i.e., the maximum cooling consumption per unit area in the same room, under the premise of equivalent external heat load; that is, the denominator corresponds to the room with the maximum internal heat load. Therefore, the meaning of the conversion coefficient k i1 is that the ratio of the internal heat load per unit area of the current room i to the room with the maximum internal heat load.

[0262] Since more cooling capacity is needed to achieve a lower indoor temperature, the comfort coefficient is introduced in this embodiment, i.e., the same room under the same other working condition, in order to achieve the current indoor temperature, relative to a preset comfortable temperature, the proportion of the amount of cooling capacity needed to be consumed or the amount of cooling capacity needed to be consumed k i2 . For example, the comfortable temperature can be taken as 25 degrees Celsius when cooling, and 18 degrees Celsius when heating.

[0263] This embodiment converts the current working condition of each room to the room with the maximum internal load based on the conversion coefficient k i1 , and cleverly avoids the influence of the heat load caused by different orientations or directions through proportional calculation; and based on the generalization of the cooling model, the comfort coefficient k i2The cooling consumption of different indoor temperatures is compensated by the calculation. Thus, the final billing can overcome the influence of room orientation and temperature setting difference. The model itself takes air temperature, six-direction adjacent room temperature and other factors as input, which reflects the difference of sunlight on different orientation rooms and the influence of adjacent rooms on cooling consumption. Through the normalization correction of the cooling equivalent of each room per unit area under the same working condition of different rooms, and the proportional correction of the cooling consumption under different target room temperatures of the same room, the differences of orientation and envelope structure, and the temperature actually enjoyed are compensated.

[0264] Embodiment 4:

[0265] In combination with Figure 1A , Figure 4A , Figure 4B and Figure 7A , Figure 7B , the embodiment provides a central air conditioning billing system 1000 based on comfort, which comprises:

[0266] a user interface unit 140 for typing parameters, initiating operations and human-computer interaction;

[0267] a driver 400 for driving the fan corresponding to the central air conditioning terminal unit;

[0268] a reference air conditioning unit 600 for pre-acquiring working characteristics as a reference for cooling capacity;

[0269] a sensing detection unit 120 for detecting the working condition and environment of the central air conditioning and the reference air conditioning unit;

[0270] a working condition adjustment unit 130 for adjusting the running working condition of the central air conditioning terminal unit and the reference air conditioning unit;

[0271] a control unit 150 connected with the driver 400, the reference air conditioning unit 600, the sensing detection unit 120, the user interface unit 140 and the working condition adjustment unit 130;

[0272] The control unit 150 is configured to:

[0273] First, the cooling equivalent of the terminal unit is modeled,

[0274] the room where the terminal unit is located is taken as the heat load, the working condition of the central air conditioning system is taken as the input quantity, and the cooling capacity equivalent of the terminal unit per unit time is taken as the output quantity, a first neural network is established in the control unit,

[0275] The cooling supply amount of the terminal unit is equivalent to the cooling supply amount of the reference air conditioning unit under the same heat load condition, and the first data sample is collected under different working conditions to train the first neural network;

[0276] Secondly, the central air conditioning cooling supply rooms are classified according to the orientation type, and a cooling consumption model is established for each type of orientation room,

[0277] The current room temperature and outdoor temperature are two scalar quantities, and the vector composed of the indoor temperature values of the current room in six directions of front, back, left, right, up and down is used as the input quantity, and the cooling consumption equivalent per unit time of the current room is used as the output quantity, and the second neural network is established in the control unit according to the room,

[0278] The terminal unit in the current room is adaptively controlled, and the room temperature is kept unchanged in stages, and when the working condition is approximately stable, the second data sample of the second neural network under different input conditions is collected, the output quantity in the sample is predicted by the first neural network, and the training is carried out based on the collected data sample;

[0279] Based on the offline established cooling supply and consumption models, the central air conditioner is charged by time period in online application,

[0280] First, according to the cooling consumption model, the correction coefficient of the current room working condition is calculated In the formula, i is the current room number, the room number i = 1 ~ N is any integer, N is the total number of rooms, p i , p i0 is the mapping output of the second neural network when the input vector is , and is the current working condition of room i, the current room temperature and outdoor temperature are the same as , and the indoor temperature values of the current room in six directions are the current room temperature;

[0281] Then the cost of the i-th room in the d-th time period is calculated as:

[0282] Wherein, PF i (t) is the cumulative cooling supply amount of the current period, Q jd is the cooling supply amount of room j in the current period, k j3 is the corresponding coefficient of room j, C d is the total cost of the central air conditioner to be allocated in the current period.

[0283] Referring to Figure 1BAs shown, when the central air conditioner is a VRF air conditioner, the driver is built in the end unit of the central air conditioner to drive the fan motor, and the control unit controls the end unit to work in a PWM mode through the driver interface.

[0284] As preferred, in the comfort-based central air conditioner billing system 1000, the cost allocation calculation is replaced by:

[0285] First, the conversion coefficient of the current room condition is calculated In the formula, i is the current room number, s i , s j are the areas of the i-th and j-th rooms, respectively, i and j are any integer in the range of 1 to N, N is the total number of rooms, p i , p j are the respective comfort coefficients of rooms i and j when the current room temperature is taken as the input vector to the second neural network, and the mapping output of the second neural network corresponding to each of them, is the current condition of room i,

[0286] and the comfort coefficient p is calculated. ic In the formula, p is the mapping output of the second neural network corresponding to room i when the current room temperature is taken as the input vector , and the difference between the conditions and is that the current room temperature is a preset comfortable temperature, such as 25 degrees Celsius;

[0287] The cost of the i-th room in the d-th time period is then calculated as:

[0288] where k j1 , k j2 are the coefficients corresponding to room j, and C d is the total cost of the central air conditioner to be allocated in the current period.

[0289] In order to realize fair charging according to the cooling capacity, the cooling capacity characteristics of the terminal unit and the cooling consumption characteristics of the rooms in different orientations are modeled respectively, and the predicted cooling capacity of the terminal unit and the cooling consumption of the room are calculated based on the two nonlinear models; the actual cooling consumption is corrected based on the cooling consumption characteristics of the rooms in different orientations to overcome the difference of the environmental influence in different orientations. In the first modeling, in order to avoid the error caused by the detection of the small air flow of the terminal unit in online application, the main pipe of the chilled water supply of the central air conditioner is taken as the measuring point, the flow and the temperature difference between the supply water and the return water are detected as the input of the air damper metering mapping model; the reference air conditioning unit is used as the calculation reference of the cooling capacity of the terminal unit through the consistency control of the thermal load condition, the nonlinear mapping model is trained by the sample set, and the cooling capacity of the terminal unit of the central air conditioner is predicted and calculated by the model obtained through the training in online application. In the second modeling, the influence of the sunlight, the maintenance structure and the adjacent room is effectively compensated through the design of the model parameter structure and the correction coefficient formula. The application can ensure the metering accuracy without increasing the cost, realize the decoupling of the mutual restriction between the terminal units in different air dampers, and accurately meter the dynamically changing cooling capacity; and through the introduction of the compensation of the actual working condition in the correction calculation of the cooling consumption of the room in different orientations, the fair and reasonable charging is ensured.

[0290] It can be understood that after the refrigeration and heating conditions are interchanged in the application, the application is also applicable to the apportionment charging of the terminal unit of the central air conditioner in the heating season.

[0291] The above describes several embodiments of the application, but these embodiments are presented as examples and do not limit the scope of the application. These embodiments can be implemented in other various ways, and various omissions, substitutions, combinations, changes can be made within the scope of the application without departing from the main idea of the application. These embodiments or their modifications are included in the scope or the main idea of the application, and are also included in the application and the equivalent scope recorded in the claims.

Claims

1. A comfort-based central air conditioning billing device, comprising a control unit, and a sensing and detection unit, a user interface unit, and an operating condition adjustment unit connected to the control unit; The sensing and detection unit detects the operating condition and environmental conditions of the central air conditioning system, and the control unit is configured to: First, a metering model is performed on the cooling equivalent of the central air conditioning terminal units. Using the room where the terminal unit is located as the heat load, the operating conditions of the central air conditioning system as the input, and the equivalent cooling capacity per unit time of the terminal unit in this room as the output, a first neural network is established in the control unit. The cooling capacity of the reference air conditioning unit in the system under the same heat load is used as the equivalent cooling capacity of the terminal unit. First data samples are collected under different operating conditions and the first neural network is trained. Secondly, the rooms cooled by central air conditioning are classified according to their location, and a cooling consumption model is established for each type of room. The system uses two scalars—the current room temperature and the outdoor temperature—along with a vector composed of the room's indoor temperature values ​​in six directions (front, back, left, right, up, and down) during cooling as input, and the equivalent cooling load per unit time for the room as output. A second neural network is then established in the control unit, organized by room. Adaptive control is performed on the terminal units in the current room to maintain the room temperature at a constant level in stages. When the operating conditions are approximately stable, second data samples of the second neural network are collected under different input conditions. The output of the samples is predicted by the first neural network, and training is performed based on the collected data samples. Based on the offline cooling and cooling consumption model, the cost of central air conditioning is allocated according to time periods when applied online. First, calculate the conversion factor based on the current room conditions. In the formula, i is the current room number, and s i s j The areas of rooms i and j are respectively, where i and j are any integers from 1 to N, and N is the total number of rooms. i p j Rooms i and j are respectively... When used as input vectors, each corresponds to the mapping output of the second neural network. The current operating status of room i. And calculate the comfort factor. In the formula, p ic For room i When used as an input vector, it corresponds to the mapping output of the second neural network, operating condition. and The only difference is that the current room temperature is a preset comfortable temperature, such as 25 degrees Celsius. The cost of room i in time period d is then calculated as follows: in, k j1 k j2 C is the coefficient corresponding to room j. d This represents the total unallocated cost of central air conditioning for this period.

2. The comfort-based central air conditioning billing device according to claim 1, characterized in that: The central air conditioning is a water-cooled central air conditioning, and the terminal unit is a terminal fan baffle. The first neural network uses a vector composed of four scalars: chilled water supply flow rate of the central air conditioning unit, supply and return water temperature difference, current temperature and humidity of the room, and the opening status values ​​of all terminal fan baffles as input. Alternatively, the central air conditioning system is an all-air central air conditioning system, the terminal unit is a terminal fan baffle, and the first neural network uses a vector composed of four scalars—the air supply flow rate of the central air conditioning unit's air outlet, the supply and return air temperature difference, the current temperature and humidity of the room, and the opening status values ​​of all terminal fan baffles—as input. Alternatively, the central air conditioning system is a variable refrigerant flow central air conditioning system, and the terminal unit is an indoor unit. The first neural network uses a vector composed of four scalars: the power of the central air conditioning unit, the power of the indoor unit, the current temperature and humidity of the room, and the on / off status values ​​of all terminal unit fans as input quantities.

3. The comfort-based central air conditioning billing device according to claim 1, characterized in that, When collecting the first data sample: A target temperature is set based on the room's initial temperature. The terminal fan driver is controlled to adjust the room temperature to the target temperature, and the current humidity is obtained as the target humidity. After the room remains at the target temperature for a period of time, the reference air conditioning unit and the terminal fan each operate independently as a cold source for a certain period to collect samples. While both cold sources are operating, the room temperature is maintained at the target temperature. The terminal fan operates in PWM mode, and the reference air conditioning unit, while operating, also maintains the humidity at the target humidity through a condition adjustment unit. Based on the operating characteristics and conditions of the reference air conditioning unit, the cooling power consumption is calculated and divided by the average PWM duty cycle of the terminal fan. The resulting value is used as the equivalent cooling capacity per unit time when the terminal fan is currently in operation (F).

4. The comfort-based central air conditioning billing device according to claim 1, characterized in that, The cost allocation for central air conditioning also includes a portion allocated by area. Therefore, the cost for room i in time period d is: Among them, C d0 This refers to the basic cost of central air conditioning.

5. The comfort-based central air conditioning billing device according to claim 1, characterized in that, The control unit is provided with a main processing module, which includes a sample extraction module for collecting data samples of the second neural network. The sample extraction module is configured to periodically and continuously collect the heat load conditions and cooling consumption parameters of the central air-conditioned cooling room, and search for the period in which the input parameter values ​​of the second neural network change within a preset fluctuation threshold, and store the data of the selected period into the data sample set of the second neural network.

6. The central air conditioning billing device based on comfort according to claim 1, characterized in that, When the central air conditioning system is a water-cooled central air conditioning system or an all-air central air conditioning system, the sensing and detection unit is equipped with a flow detection module and a temperature detection module located at the inlet of the cooling main pipe of the central air conditioning unit, and a temperature detection module at the outlet of the return main pipe connected to the unit. Multiple temperature detection modules located in different positions within the room are positioned at the same height and on two perpendicular diagonal lines. The humidity is sensed by a humidity detection module located in the middle of the room's return air duct within the sensing unit. Corresponding to the low, medium, and high speed settings of the fan, the fan speed opening value of the terminal unit fan can be taken as the normalized power value of the three fans corresponding to each of the three speed settings.

7. The comfort-based central air conditioning billing device according to claim 1, characterized in that, The operating condition adjustment unit includes a temperature uniformity module, which comprises a base, a vertical rotating shaft, a curved support arm, a horizontal rotating shaft, and a tiltable bracket with two sections of support arms movably connected by bolts. A telescopic support rod forming an acute angle with the axis of the horizontal rotating shaft is connected between the outer ends of the two support arms. A temperature uniformity fan is mounted at the end of the tiltable bracket. The control unit is also configured to control the operation of the temperature uniform module, so that the temperature uniform fan shaft moves in a spatial spiral to deliver the cold air blown out by the central air conditioning terminal unit and / or the reference air conditioning unit to all directions of the room until the temperature difference of the plurality of temperature detection modules in the room is less than the temperature difference threshold.

8. A comfort-based central air conditioning billing system, comprising: A user interface unit for inputting parameters, initiating operations, and human-computer interaction; A reference air conditioning unit whose operating characteristics have been pre-acquired and used as a reference for cooling capacity; A sensor detection unit used to detect the operating conditions and environment of central air conditioning and reference air conditioning units; Operating condition adjustment unit used to adjust the operating conditions of central air conditioning terminal units and reference air conditioning units; A control unit connected to the reference air conditioning unit, the sensing and detection unit, the user interface unit, and the operating condition adjustment unit; wherein, the control unit is configured as follows: First, a metering model is performed on the cooling equivalent of the terminal unit. Using the room where the terminal unit is located as the heat load, the operating conditions of the central air conditioning system as the input, and the equivalent cooling capacity per unit time of the terminal unit in this room as the output, a first neural network is established in the control unit. The cooling capacity of the reference air conditioning unit in the system under the same heat load is used as the equivalent cooling capacity of the terminal unit. First data samples are collected under different operating conditions and the first neural network is trained. Secondly, the rooms cooled by central air conditioning are classified according to their location, and a cooling consumption model is established for each type of room. The system uses two scalars—the current room temperature and the outdoor temperature—along with a vector composed of the room's indoor temperature values ​​in six directions (front, back, left, right, up, and down) during cooling as input, and the equivalent cooling load per unit time for the room as output. A second neural network is then established in the control unit, organized by room. Adaptive control is performed on the terminal units in the current room to maintain the room temperature at a constant level in stages. When the operating conditions are approximately stable, second data samples of the second neural network are collected under different input conditions. The output of the samples is predicted by the first neural network, and training is performed based on the collected data samples. Based on the offline cooling and cooling consumption model, the cost of central air conditioning is allocated according to time periods when applied online. First, calculate the conversion factor based on the current room conditions. In the formula, i is the current room number, and s i s j The areas of rooms i and j are respectively, where i and j are any integers from 1 to N, and N is the total number of rooms. i p j Rooms i and j are respectively... as input vector Each of them corresponds to the mapping output of the second neural network. The current operating status of room i. And calculate the comfort factor. In the formula, p ic For room i as input vector At that time, it corresponds to the mapping output of the second neural network, working condition and The only difference is that the current room temperature is a preset comfortable temperature, such as 25 degrees Celsius. The cost of room i in time period d is then calculated as follows: Where, k j1 k j2 C is the coefficient corresponding to room j. d This represents the total unallocated cost of central air conditioning for this period.

9. The comfort-based central air conditioning billing system according to claim 8, characterized in that, The sample data for the first neural network are collected in the following manner: After the room temperature is maintained at the target temperature for a period of time, the drive is turned off and a timer is started. The reference air conditioning unit is controlled to maintain the room temperature at the target temperature from t=0 to t=T1 in the first time period. At the same time, the operating condition adjustment unit is controlled to maintain the room humidity at the target humidity. The cooling power q(t) is calculated based on the operating characteristics and operating conditions of the reference air conditioning unit, and the cooling capacity of the first time period is accumulated. After shutting down the reference air conditioning unit and timing again, and setting the fan speed on / off value F, the driver is controlled to operate in PWM mode to maintain the room temperature at the target temperature from t=0 to t=T2 in the second time period, and its equivalent time is calculated. Where Δ(t) is the PWM value, The drive is turned off again and the timing is restarted. The reference air conditioning unit is controlled to maintain the room temperature at the target temperature during the third time period from t=0 to t=T1. At the same time, the operating condition adjustment unit is controlled to maintain the room humidity at the target humidity. The cooling capacity for the third time period is calculated again. Furthermore, the reference air conditioning unit employs a heating and cooling air conditioner; wherein, if the ratio of the equivalent time dT to the duration T2 of the second time period is less than the duty cycle threshold Δs, during the data acquisition, within the second time period, the reference air conditioning unit is also controlled to operate in heating mode from τ=0 to τ=T3, and the heating equivalent is recorded. Accordingly, the equivalent cooling capacity per unit time of the fan baffle under the current operating conditions is calculated:

10. The comfort-based central air conditioning billing system according to claim 8, characterized in that: The six-way room temperature value in the second neural network can be replaced with the on / off status value of the terminal unit fan in the six-way room; a solar radiation intensity parameter can also be added to the input of the second neural network.

Images