Variable air volume control method and device based on Kalman filter

Abstract

The present invention provides a variable air volume control method and device based on a Kalman filter. Based on a tetrahedral grid of a preset size, a target space is divided into n calculation areas; when m temperature-measurable calculation areas meet preset conditions, the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k are input into a state estimation model, and the real temperature increment matrix of the m temperature-measurable calculation areas is output; based on the real temperature increment matrix, the prior estimation matrix of the n calculation areas and the gain matrix of the Kalman filter, the posterior estimation matrix of the n calculation areas at the current time k is determined; based on the constraints of the designed air supply speed and the air supply speed increment when the air supply outlet and the return air outlet transmit air, the posterior estimation matrix and the designed temperature matrix of the n calculation areas, the air supply speed when the air supply outlet and the return air outlet transmit air in the target space is controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of heating, ventilation and air conditioning, and in particular to a variable air volume control method and a device thereof based on a Kalman filter. Background Art

[0002] With the continuous increase in the scale and functional complexity of modern buildings, large-space buildings, such as airport terminals, station waiting halls, exhibition centers, industrial plants, etc., have increasingly stringent requirements for indoor environmental control. In these large-space buildings, the design and operation of the heating, ventilation and air conditioning (HVAC) system is not only directly related to the comfort of the indoor environment, but also has a profound impact on the overall energy efficiency and operating costs of the building. However, the thermal environment of large-space buildings presents significant spatiotemporal non-uniform characteristics, with large spatial volumes, uneven heat load distribution and dynamic changes over time, and complex coupling between the temperature field and the airflow field. It is difficult to effectively achieve variable air volume control through traditional empirical models or fixed parameters. Summary of the invention

[0003] In view of this, the present invention provides a variable air volume control method and device based on a Kalman filter.

[0004] One aspect of the present invention provides a variable air volume control method based on a Kalman filter, comprising: dividing a target space into n calculation areas based on a tetrahedral grid of a preset size, wherein the target space contains an air supply port and an air return port, the air supply port is used to transmit air from outside the target space to inside the target space, and the air return port is used to transmit air from inside the target space to outside the target space, each calculation area includes at least one tetrahedral grid, n is a positive integer, and there are m temperature-measurable calculation areas in the n calculation areas, m≤n; when it is determined that the m temperature-measurable calculation areas meet the preset conditions, the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k are input into a state estimation model, and the m temperature-measurable a real temperature increment matrix of the calculation area, wherein the state estimation model is constructed based on the Kalman filter; based on the real temperature increment matrices of the m calculation areas with measurable temperatures, the prior estimation matrices of the n calculation areas and the gain matrix of the Kalman filter, the posterior estimation matrices of the n calculation areas at the current time k are determined, wherein the gain matrix of the Kalman filter is determined based on the covariance matrix of the process noise and the covariance matrix of the measurement noise; and based on the constraints on the designed air supply speed and air supply speed increment when the supply and return air outlets transmit air, the posterior estimation matrices of the n calculation areas at the current time k and the designed temperature matrix of the n calculation areas, the air supply speed when the supply and return air outlets in the target space transmit air is controlled.

[0005] Another aspect of the present invention provides a variable air volume control device based on a Kalman filter, comprising: a division module, for dividing a target space into n calculation areas based on a tetrahedral grid of a preset size, wherein the target space contains an air supply port and an air return port, the air supply port is used to transmit air from outside the target space to inside the target space, and the air return port is used to transmit air from inside the target space to outside the target space, each calculation area includes at least one tetrahedral grid, n is a positive integer, and there are m temperature-measurable calculation areas in the n calculation areas, m≤n; a processing module, for inputting the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k into a state estimation model when it is determined that the m temperature-measurable calculation areas meet the preset conditions, and outputting the m temperature-measurable a real temperature increment matrix of the calculation area, wherein the state estimation model is constructed based on the Kalman filter; a determination module, for determining the posterior estimation matrix of the n calculation areas at the current time k based on the real temperature increment matrix of the m temperature-measurable calculation areas, the prior estimation matrix of the n calculation areas and the gain matrix of the Kalman filter, wherein the gain matrix of the Kalman filter is determined based on the covariance matrix of the process noise and the covariance matrix of the measurement noise; and a control module, for controlling the supply air speed when the supply air outlet and the return air outlet transmit air in the target space based on the constraints on the designed supply air speed and the supply air speed increment when the supply air outlet and the return air outlet transmit air, the posterior estimation matrix of the n calculation areas at the current time k and the designed temperature matrix of the n calculation areas.

[0006] According to an embodiment of the present invention, a state estimation model is constructed based on a Kalman filter. When it is determined that m temperature-measurable calculation areas meet preset conditions, a posterior estimation matrix is ​​determined based on a real temperature increment matrix determined by measurement noise, a gain matrix of a Kalman filter, and a priori estimation matrix, so as to control the air supply speed when the air supply outlet and the return air outlet in the target space transmit air. Since the gain matrix of the Kalman filter is determined based on the covariance matrix of process noise and measurement noise, the posterior error of the posterior estimation can be minimized by the gain matrix of the Kalman filter, and a high-precision temperature increment estimation is provided under complex working conditions, and the temperature change can be responded to in real time, so as to accurately control the air supply speed. Since the area of ​​the air supply outlet or the return air outlet in the target area is fixed, the air supply speed is controlled based on the temperature change responded in real time, and the air supply volume can be dynamically controlled to realize variable air volume control. In addition, by determining the posterior estimation matrix of the calculation area, it is not necessary to install temperature measurement equipment in all the calculation areas, so that the energy loss of excessive installation of temperature measurement equipment can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The above and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings.

[0008] Figure 1 A flow chart of a variable air volume control method based on a Kalman filter according to an embodiment of the present invention is shown.

[0009] Figure 2 A schematic plan view of a calculation area for target space division according to an embodiment of the present invention is shown.

[0010] Figure 3 A flow chart of determining a gain matrix of a Kalman filter according to an embodiment of the present invention is shown.

[0011] Figure 4 A schematic diagram showing the relationship between the air volume and air supply speed for each of the four interfaces of the tetrahedral grid according to an embodiment of the present invention is shown.

[0012] Figure 5 A schematic diagram is shown of controlling the air supply speed when air is transmitted through the air supply outlet and the return air outlet in the target space according to an embodiment of the present invention.

[0013] Figure 6 A block diagram of a variable air volume control device based on a Kalman filter according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0014] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0015] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0016] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0017] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0018] In the process of realizing the concept of the present invention, it was found that when a constant air volume or a simple proportional integral differential (PID) control strategy is used to control the air supply volume in a large space environment, there are problems of insufficient dynamic adaptability and insufficient control accuracy. Model Predictive Control (MPC), as a feedforward optimization control strategy, can adjust the control strategy in real time to adapt to the dynamic changes of the system, but it is not well adapted to the control of air supply volume in large spaces. This is because the airflow and temperature distribution in a large space involves a large number of state variables and input and output parameters. Directly using a global model for real-time prediction will significantly increase the computational complexity and make it difficult to meet real-time requirements. In addition, due to the number and location of sensor arrangements, the airflow and temperature distribution in a large space are difficult to be fully observed, resulting in low control accuracy.

[0019] Based on this, an embodiment of the present invention provides a variable air volume control method based on a Kalman filter and a device thereof. The method includes: based on a tetrahedral grid of a preset size, dividing the target space into n calculation areas, wherein the target space contains an air supply port and an air return port, the air supply port is used to transmit air from outside the target space to inside the target space, and the air return port is used to transmit air from inside the target space to outside the target space, each calculation area includes at least one tetrahedral grid, n is a positive integer, and there are m temperature-measurable calculation areas in the n calculation areas, m≤n; when it is determined that the m temperature-measurable calculation areas meet the preset conditions, the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k are input into the state estimation model, and the real temperature increment matrix of the m temperature-measurable calculation areas is output, Among them, the state estimation model is constructed based on the Kalman filter; based on the real temperature increment matrix of m temperature-measurable calculation areas, the prior estimation matrix of n calculation areas and the gain matrix of the Kalman filter, the posterior estimation matrix of the n calculation areas at the current time k is determined, wherein the gain matrix of the Kalman filter is determined based on the covariance matrix of the process noise and the covariance matrix of the measurement noise; and based on the constraints on the designed supply air speed and supply air speed increment when the supply air outlet and the return air outlet transmit air, the posterior estimation matrix of the n calculation areas at the current time k and the designed temperature matrix of the n calculation areas, the supply air speed when the supply air outlet and the return air outlet transmit air in the target space is controlled.

[0020] The following will be passed Figure 1~Figure 5 The variable air volume control method based on the Kalman filter according to the embodiment of the present invention is described in detail.

[0021] Figure 1 A flow chart of a variable air volume control method based on a Kalman filter according to an embodiment of the present invention is shown.

[0022] like Figure 1 As shown, the variable air volume control method 100 based on the Kalman filter includes operations S110 to S140.

[0023] In operation S110 , the target space is divided into n calculation regions based on a tetrahedral grid of a preset size.

[0024] In operation S120, when it is determined that the m temperature-measurable calculation areas meet the preset conditions, the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k are input into the state estimation model, and the real temperature increment matrix of the m temperature-measurable calculation areas is output.

[0025] In operation S130, based on the real temperature increment matrices of the m temperature-measurable calculation regions, the prior estimation matrices of the n calculation regions, and the gain matrix of the Kalman filter, the a posteriori estimation matrices of the n calculation regions at the current time k are determined.

[0026] In operation S140, the supply air speed when the supply air outlet and the return air outlet transport air is controlled based on the constraints on the designed supply air speed and the supply air speed increment when the supply air outlet and the return air outlet transport air, the a posteriori estimation matrix of the n calculation areas at the current time k, and the design temperature matrix of the n calculation areas.

[0027] According to an embodiment of the present invention, the preset size can be determined based on software capable of tetrahedral meshing. The tetrahedral mesh may include four interfaces: upper, lower, left, and right. The target space may contain an air supply outlet and a return air outlet. The air supply outlet can be used to transfer air from outside the target space to inside the target space, and the return air outlet can be used to transfer air from inside the target space to outside the target space. Each calculation area may include at least one tetrahedral mesh, n is a positive integer, and there are m temperature-measurable calculation areas in n calculation areas, m≤n. Each calculation area may contain an air supply outlet, may contain a return air outlet, or may contain neither an air supply outlet nor a return air outlet.

[0028] Exemplarily, the target space may be meshed based on the geometric model of the target space to obtain a plurality of tetrahedral meshes of preset sizes, and the tetrahedral meshes in the plurality of tetrahedral meshes of preset sizes may be meshed to obtain n calculation regions.

[0029] According to an embodiment of the present invention, a preset condition can be used to characterize the conditions for estimating the temperature value of a temperature-unmeasurable calculation area using the measured value of a temperature-measurable calculation area. A measured temperature noise matrix can be used to indicate the temperature measurement noise of each temperature-measurable calculation area in m temperature-measurable calculation areas. A measured temperature increment matrix can be used to represent the measured value of the temperature increment of each calculation area in m temperature-measurable calculation areas, and can also be used to represent the measured value of the temperature increment of each calculation area in n calculation areas. The measured value of the temperature increment for a temperature-unmeasurable calculation area can be zero. The temperature increment can be used to indicate the temperature change at the current moment k relative to the previous moment k-1 at the current moment k. The measured values ​​of the temperature increment of the m temperature-measurable calculation areas can be acquired by a temperature sensor.

[0030] According to an embodiment of the present invention, the real temperature increment matrix can be used to indicate the real value of the temperature increment. The state estimation model can be constructed based on a Kalman filter. The state estimation model can be used to determine the real value of the temperature increment of each calculation area in the m temperature-measurable calculation areas.

[0031] Exemplarily, the state estimation model can sum the temperature measurement noise of each temperature-measurable calculation area and the measured value of the temperature increment of each calculation area in the m temperature-measurable calculation areas to obtain the true value of the temperature increment of each calculation area in the m temperature-measurable calculation areas.

[0032] According to an embodiment of the present invention, the gain matrix of the Kalman filter can be determined based on the covariance matrix of the process noise and the covariance matrix of the measurement noise. The gain matrix of the Kalman filter can be used to minimize the a posteriori error between the a posteriori estimate value and the true value at the current time k.

[0033] For example, at the current time k, the product of the prior estimation matrix of the n calculation regions and the screening matrix of the m temperature-measurable calculation regions can be determined first, and then the difference between the real temperature increment matrix of the m temperature-measurable calculation regions and the product can be determined. The difference is multiplied by the gain matrix of the Kalman filter and then summed with the prior estimation matrix of the n calculation regions to obtain the posterior estimation matrix of the n calculation regions.

[0034] According to an embodiment of the present invention, the constraints of the air supply speed designed when the air supply outlet and the return air outlet transmit air can be the same, such as the maximum constraint on the air supply speed and the minimum constraint on the air supply speed. The constraints of the air supply speed increment designed when the air supply outlet and the return air outlet transmit air can be the same, such as including the maximum constraint on the air supply speed increment and the minimum constraint on the air supply speed increment. The a posteriori estimation matrix is ​​used to indicate the a posteriori estimation value of the temperature increment. The design temperature matrix is ​​used to indicate the design temperature value. The design temperature value can be determined according to the temperature requirement in the target space.

[0035] For example, based on the constraint conditions, the air supply speed of the air supply and return air outlets in the target space when transmitting air can be changed so that the difference between the posterior estimation value of each calculation area and the design temperature value of each calculation area is minimized, and the air supply speed of the air supply and return air outlets in the target space when the difference is minimized is used as the optimized air supply speed, and the optimized air supply volume is obtained according to the product of the optimized air supply speed and the area of ​​each air supply or return air outlet. The air volume is controlled according to the optimized air supply volume.

[0036] According to an embodiment of the present invention, a state estimation model is constructed based on a Kalman filter. When it is determined that m temperature-measurable calculation areas meet preset conditions, a posterior estimation matrix is ​​determined based on a real temperature increment matrix determined by measurement noise, a gain matrix of a Kalman filter, and a priori estimation matrix, so as to control the air supply speed when the air supply outlet and the return air outlet in the target space transmit air. Since the gain matrix of the Kalman filter is determined based on the covariance matrix of process noise and measurement noise, the posterior error of the posterior estimation can be minimized by the gain matrix of the Kalman filter, and a high-precision temperature increment estimation is provided under complex working conditions, and the temperature change can be responded to in real time, so as to accurately control the air supply speed. Since the area of ​​the air supply outlet or the return air outlet in the target area is fixed, the air supply speed is controlled based on the temperature change responded in real time, and the air supply volume can be dynamically controlled to realize variable air volume control. In addition, by determining the posterior estimation matrix of the calculation area, it is not necessary to install temperature measurement equipment in all the calculation areas, so that the energy loss of excessive installation of temperature measurement equipment can be reduced.

[0037] According to an embodiment of the present invention, the target space may include at least one of the following: an airport terminal, an exhibition center, an industrial plant, and a station. The station may include a station platform, such as a subway platform, a high-speed rail platform, etc. Whether it is a scene with frequent load changes or an area with large external environmental disturbances, the present invention can achieve precise control through variable air volume control.

[0038] Figure 2 A schematic plan view of a calculation area for target space division according to an embodiment of the present invention is shown.

[0039] According to an embodiment of the present invention, operation S110 may include the following operations: using a modeling tool to establish a geometric model of the target space, using a tetrahedral mesh of a preset size in a tetrahedral meshing software to mesh the target space, and obtaining a plurality of tetrahedral meshes of a preset size. The target space may be a closed space having an air supply vent and an air return vent. For example, taking the target space as a subway platform, the waiting area of ​​the subway platform is 151.3 meters long, 11.4 meters wide, and 3 meters high, and the subway platform may evenly distribute 28 air supply vents and 22 air return vents.

[0040] Multiple tetrahedral meshes of preset sizes can be grouped together as a calculation area. A calculation area contains at most one air supply outlet or one air return outlet. The target space can be divided in this way to obtain multiple calculation areas, such as Figure 2 In addition, the blank squares in the gray squares refer to the wall columns in the target space, and the solid black dots in the gray squares refer to the air supply or return outlets.

[0041] According to an embodiment of the present invention, by introducing a tetrahedral grid of a preset size and then dividing the calculation area, the calculation complexity can be reduced.

[0042] According to an embodiment of the present invention, a screening matrix of m temperature measurable calculation areas can be determined according to the calculation area of ​​measurable temperature: , and determine the matrix rows Is it full rank? If it is full rank, it is determined that the m temperature-measurable calculation areas meet the preset conditions, and the temperature values ​​of all calculation areas can be estimated based on the temperature measurement values ​​of the m temperature-measurable calculation areas. If it is not full rank, it can be updated by installing additional temperature measurement equipment. , until the matrix has full row rank.

[0043] Figure 3 A flow chart of determining a gain matrix of a Kalman filter according to an embodiment of the present invention is shown.

[0044] According to an embodiment of the present invention, the measured temperature increment matrix can be used to indicate the measured value of the temperature increment, the real temperature increment matrix can be used to indicate the real value of the temperature increment, and the temperature increment is used to indicate the temperature change at the current moment k relative to the previous moment k-1 of the current moment k.

[0045] In addition to the above, variable air volume control methods may include Figure 1 In addition to the operations S110 to S140 shown in FIG. Figure 3 Operations S310 to S340 are shown.

[0046] In operation S310, a covariance matrix of a posterior error is determined according to a posterior error between a posterior estimated value and a true value at a current time k.

[0047] In operation S320, the trace of the covariance matrix of the posterior error is differentiated with respect to the gain matrix of the Kalman filter, and the derivative is set to zero, thereby obtaining an equation of the gain matrix of the Kalman filter with respect to the covariance matrix of the a priori error at the current time k and the covariance matrix of the measurement noise.

[0048] In operation S330, a covariance matrix of a priori errors at a current time k is determined based on a covariance matrix of process noise and a covariance matrix of a posteriori errors at a previous time k-1.

[0049] In operation S340, a gain matrix of the Kalman filter is determined based on the covariance matrix of the a priori error at the current time k, the covariance matrix of the measurement noise, and the equation.

[0050] According to an embodiment of the present invention, the covariance matrix of the a posteriori error of the previous moment k-1 can be determined based on the a posteriori error of the a posteriori estimate of the previous moment k-1. The covariance matrix of the process noise can be determined based on the temperature increment of the n calculation areas calculated based on the state estimation model and the fluid dynamics model. The covariance matrix of the measurement noise can be determined based on the measurement noise of the temperature sensor. It should be noted that if the current moment k is the initial moment, the a posteriori estimate of the previous moment k-1 can be an empirical value.

[0051] According to an embodiment of the present invention, a Kalman filter is used to estimate the temperature field in real time, and the gain matrix of the Kalman filter can be used to minimize the error of the posterior estimation value, that is, the posterior error between the posterior estimation value and the true value is minimized, and the posterior error obeys a Gaussian distribution, so the calculated posterior error is minimized, that is, the trace of the posterior error covariance matrix is ​​minimized. Thus, high-precision temperature increment estimation is provided under complex working conditions.

[0052] According to an embodiment of the present invention, the a priori estimation matrix may be used to indicate a priori estimation values ​​of the temperature increment, and the a priori estimation values ​​may be determined based on a posterior estimation value at the previous moment k-1.

[0053] In addition to the above, variable air volume control methods may include Figure 1 Operation S110 to operation S140 shown in FIG. Figure 3In addition to the operations S310 to S340 shown, the following operations may also be included: determining the temperature increments of the n calculation areas calculated by the state estimation model based on the prior estimation matrix of the n calculation areas; using the air supply speed when the supply air outlet and the return air outlet transmit air as a variable, simulating the calculation using the fluid dynamics model to obtain the temperature increments of the n calculation areas calculated by the fluid dynamics model; determining the covariance matrix of the process noise based on the covariance of the temperature increments of the n calculation areas calculated by the state estimation model and the temperature increments of the n calculation areas calculated by the fluid dynamics model; and constructing a diagonal matrix based on the square of the measurement noise of the temperature sensor to obtain the covariance matrix of the measurement noise.

[0054] According to an embodiment of the present invention, process noise and measurement noise conform to Gaussian distribution, and process noise is caused by temperature calculation errors of the target space temperature field distribution, and the temperature calculation errors of each calculation area are interrelated. Therefore, the covariance matrix of process noise is generally complex and difficult to obtain. Therefore, the present invention determines the covariance matrix of process noise based on the covariance of the temperature increments of n calculation areas calculated by the state estimation model and the temperature increments of n calculation areas calculated by the fluid dynamics model. This can overcome the problem that the covariance matrix of process noise is complex and difficult to obtain, and accurately determine the covariance matrix of process noise.

[0055] For example, process noise Obey the probability distribution like , Represents the covariance matrix of process noise, with dimension ; Represents the temperature increment of n calculation areas calculated by the fluid dynamics model, with dimensions of , the unit is ; Represents the temperature increment of n computational regions calculated by the state estimation model, with dimensions of , the unit is , The function refers to the covariance function.

[0056] The measurement noise is caused by the measurement noise of the temperature measurement device such as the temperature sensor corresponding to the calculation area of ​​the measurable temperature. Therefore, the measurement noises in each area are independent of each other, and the covariance matrix of the measurement noise is a diagonal matrix. Therefore, a diagonal matrix is ​​constructed according to the square of the measurement noise of the temperature sensor, and the covariance matrix of the measurement noise can be accurately obtained.

[0057] For example, measuring noise Obey the probability distribution like , Represents the covariance matrix of the measurement noise, with dimension ; Represents the measurement noise of the temperature sensor, with dimension , the unit is .

[0058] According to an embodiment of the present invention, the equation may be as shown in the following equation (1):

[0059] (1)

[0060] Among them, K(k) represents the gain matrix of the Kalman filter at the current moment k, and its dimension is ; Represents the screening matrix of m temperature-measurable computational regions, with dimensions ; Represents the covariance matrix of the measurement noise, with dimension ; Represents the covariance matrix of the prior error at the current moment k, with dimension , n represents the number of calculation regions, m represents the number of calculation regions with measurable temperature, and T represents transpose.

[0061] According to an embodiment of the present invention, the covariance matrix of the prior error at the current time k may be as shown in the following formula (2):

[0062] (2)

[0063] in, Represents the covariance matrix of the posterior error of the previous moment k-1, with dimension ; Represents the covariance matrix of process noise, with dimension , n represents the number of calculation regions, m represents the number of calculation regions with measurable temperature, and T represents transposition; Represents the coefficient matrix of the temperature variable, with dimension .

[0064] According to an embodiment of the present invention, the posterior estimation matrix of the n calculation regions at the current time k may be as shown in the following formula (3):

[0065] (3)

[0066] in, Represents the posterior estimation matrix of n computational regions at the current time k, with a dimension of ; represents the prior estimation matrix of n computational regions at the current time k, ; K(k) represents the gain matrix of the Kalman filter at the current moment k, and its dimension is ; Represents the real temperature increment matrix of m temperature-measurable calculation areas, with dimension ; The screening matrix representing the computational region where the temperature can be measured has dimensions .

[0067] For example, the prior error of the current time k is and the prior error at the current time k The covariance matrix of They can be expressed as the following equations (4) and (5) respectively:

[0068] (4)

[0069] (5)

[0070] in, Indicates the true value of the temperature increment of n calculation areas at the current time k.

[0071] The posterior error at the current time k and the posterior error at the current time k The covariance matrix of They can be expressed as follows:

[0072] (6)

[0073] (7)

[0074] For example, considering the process noise and measurement noise and the gain matrix of the Kalman filter, the posterior error at the current time k is And the covariance matrix of the posterior error at the current time k They can be expressed as follows:

[0075] (8)

[0076] (9)

[0077] in, represents the unit matrix, with dimension .

[0078] The prior error of k at the current moment It can be expressed as the following formula (10):

[0079] (10)

[0080] The trace of the covariance matrix of the posterior error at the current time k can be differentiated with respect to the Kalman gain matrix, and the derivative is set to 0, so as to obtain the Kalman gain matrix that minimizes the trace of the covariance matrix of the posterior error at the current time k. The process is shown in the following equation (11):

[0081] (11)

[0082] According to an embodiment of the present invention, the process noise covariance matrix and the measurement noise covariance matrix are used, and the Kalman gain matrix is ​​updated in real time to dynamically adjust the weights of the estimated value and the measured value to generate the optimal state estimate, and the posterior estimate of the temperature distribution and the error covariance matrix are recursively updated to ensure dynamic monitoring and high-precision control of the temperature field in the target space.

[0083] Figure 4 A schematic diagram showing the relationship between the air volume and air supply speed for each of the four interfaces of the tetrahedral grid according to an embodiment of the present invention is shown.

[0084] According to an embodiment of the present invention, the variable air volume control method may include the above Figure 1 Operation S110 to operation S140 shown in FIG. Figure 3 In addition to the operations S310 to S340 shown, the following operations may also be included: simulating the air flow in the target space with the air supply speed and preset size when the air is transmitted by the supply or return outlet as variables to obtain simulation results, wherein the simulation results are used to characterize the air supply volume of each mesh surface of the tetrahedral mesh at the preset size and air supply speed; determining the fitting relationship between the regional interface air volume and the air supply speed for the calculation area based on the simulation results, wherein the regional interface air volume is used to characterize the sum of the air supply volumes of each mesh surface in the calculation area; constructing an energy balance equation for each calculation area based on the fitting relationship and the geometric position of each calculation area in the target space; linearizing the energy balance equation, and converting the linear terms after the linearization transformation into incremental form to obtain an incremental energy balance equation; converting the incremental energy balance equation into a time discrete equation based on the time step between the current moment k and the previous moment k-1; and determining the coefficient matrix of the temperature variable based on the time discrete equation.

[0085] According to an embodiment of the present invention, in a calculation area containing an air supply or return outlet, the air flow in the target space can be simulated by changing the air supply speed and preset size of the air supply or return outlet when transmitting air. The relationship between the air volume and the air supply speed of the four interfaces of each tetrahedral mesh in the calculation area under fitting tetrahedral meshes of different sizes and different air supply speeds can be obtained by taking a certain calculation area as an example. Figure 4The relationship between the air volume and air supply speed of each of the four interfaces of a tetrahedral mesh, such as the upper interface, the lower interface, the left interface, and the right interface, shows that the air volume and air supply speed of each interface are linearly related. Based on this, the fitting relationship between the regional interface air volume and air supply speed for the calculation area can be, for example, as shown in the following formula (12):

[0086] (12)

[0087] in, Indicates the regional interface air volume in the calculation area, in units of ; Indicates the air supply speed of the air supply or return outlet, in units of ; They respectively represent the linear coefficient and constant term of the fitting relationship between air volume and air supply speed on each interface in the calculation area.

[0088] For example, the convection heat transfer effect between the wall surface and the air in the target space and the heat dissipation of people in the calculation area can be used as unpredictable disturbances for the prediction and control of the air volume in the target space. The geometric position of each calculation area in the target space can be determined in the form of row and column coordinates. For example, for a certain geometric position The calculation area can be expressed as the geometric position of the calculation area in the target space is the i-th row and the j-th column. The energy balance equation can be expressed as follows (13):

[0089] (13)

[0090] in, Indicates the calculation time, in units of , the current time k can be any calculation time; The geometric position is The volume of the calculation area is in ; express The geometric position at the moment is The derivative of the temperature in the computational region, in units of ; express The geometric position at the moment is The temperature of the calculation area is in ; express The geometric position at the moment is The temperature of the calculation area is in ; express The geometric position at the moment is The temperature of the calculation area is in ; express The geometric position at the moment is The temperature of the calculation area is in ; express The geometric position at the moment is The temperature of the calculation area is in ; Respectively represent the geometric positions The linear coefficient and constant term of the fitting relationship between the upper interface air volume and air supply speed in the calculation area; Respectively represent the geometric positions The linear coefficient and constant term of the fitting relationship between the air volume and air supply speed at the lower interface of the calculation area; Respectively represent the geometric positions The linear coefficient and constant term of the fitting relationship between the air volume and air supply speed at the left interface of the calculation area; Respectively represent the geometric positions The linear coefficient and constant term of the fitting relationship between the right interface air volume and air supply speed in the calculation area; express Air supply speed at any time, in ; Indicates the air outlet area of ​​the calculation area, in units of ; Indicates the supply air temperature of the calculation area, in units of .

[0091] It should be noted that when a certain interface in the calculation area is a wall, or the air volume is negative, or there is no air supply or return outlet in the calculation area, the corresponding item in the energy balance equation is removed. When the calculation time is the initial time, the initial value of the temperature of the calculation area and the initial value of the air supply speed can be determined according to the steady-state value of the temperature and the steady-state value of the air supply speed under the design condition.

[0092] For the nonlinear term obtained by multiplying the air supply speed and the temperature of the calculation area in formula (13), the first-order Taylor formula can be used to linearize it. For the linear term, it can be directly written in incremental form, ignoring the high-order infinitesimal terms, and the incremental energy balance equation is obtained as shown in formula (14):

[0093] (14)

[0094] in, express The geometric position at the moment is The temperature derivative of the calculation area is ; express The geometric position at the moment is The temperature change in the calculation area is ; express The geometric position at the moment is The temperature change in the calculation area is ; express The geometric position at the moment is The temperature change in the calculation area is ; express The geometric position at the moment is The temperature change in the calculation area is ; express The geometric position at the moment is The temperature change in the calculation area is ; express The geometric position at the moment is The change in air supply speed in the calculation area is in ; Indicates the initial value of the supply air temperature in units of ; Indicates the initial value of the air supply speed, in units of .

[0095] For target space The energy balance equation of all calculation areas in the target space can be used to obtain the matrix form as shown in formula (15):

[0096] (15)

[0097] in, Represents the temperature variation derivative matrix of n calculation regions in the target space, with dimensions of , the unit is ; Represents the coefficient matrix of the target space temperature variable, with dimensions ; Represents the temperature change matrix of n calculation areas in the target space, with dimensions of , the unit is ; Represents the target space input coefficient matrix, with dimension .

[0098] The non-steady-state term in equation (15) can be discretized using analytical methods, as shown in equation (16):

[0099] (16)

[0100] The time step between the current time k and the previous time k-1 can be The above equation (16) can be integrated from 0 to k+1, assuming that the air supply speed is a step function and the air supply speed increment from k to k+1 remains unchanged, which is , then the above formula (16) is transformed into the following formula (17):

[0101] (17)

[0102] The above equation (17) can be transformed into a time discrete equation, as shown in the following equation (18):

[0103] (18)

[0104] in, , , express The temperature change matrix of the target space at the moment has the dimension , the unit is ; Represents the coefficient matrix of the temperature variable, with dimension ; express The temperature change matrix of the target space at the moment has the dimension , the unit is ; Represents the state-space model input coefficient matrix, with dimension ; express The change in air supply speed of the target space at the moment, in units of .

[0105] Taking the temperature of each calculation area as the output, the establishment of the state estimation model for variable air volume control can be shown as follows (19):

[0106] (19)

[0107] in, express The temperature matrix of the target space at the moment, the dimension is , the unit is ; express The temperature matrix of the target space at the moment, the dimension is , the unit is .

[0108] According to an embodiment of the present invention, the constructed state estimation model can reflect the key characteristics of the dynamic changes in temperature in the target space environment. Compared with the traditional static model, the state space model has stronger dynamic adaptability and can respond to changes in heat load and boundary conditions in real time.

[0109] According to another embodiment of the present invention, considering the model process noise and measurement noise, a Kalman filter may be used to construct a state estimation model. The constructed state estimation model may be shown in the following equation (20):

[0110] (20)

[0111] in, Indicates the change in air supply speed of the target space at time k-1, in units of ; represents the process noise of the state estimation model at time k-1, and its dimension is ; represents the measurement noise of the state estimation model at time k, and its dimension is .

[0112] It should be noted that Only the values ​​of the rows corresponding to the calculation area of ​​measurable temperature are credible, and the values ​​of the other rows are all 0; satisfying the relationship .

[0113] Figure 5 A schematic diagram is shown of controlling the air supply speed when air is transmitted through the air supply outlet and the return air outlet in the target space according to an embodiment of the present invention.

[0114] According to an embodiment of the present invention, the a posteriori estimation matrix may be used to indicate the a posteriori estimation value of the temperature increment. Figure 5 As shown, controlling the air supply speed when the air supply outlet and the return air outlet in the target space transmit air may include operations S531 to S536.

[0115] In operation S531, based on the posterior estimation matrix of the n computational regions at the current time k, the posterior estimation value of each of the n computational regions at the current time k is determined.

[0116] In operation S532 , based on the design temperature matrix of the n calculation regions, design temperature values ​​of the n calculation regions are determined.

[0117] In operation S533, an objective function is determined according to minimizing the difference between the posterior estimation value and the design temperature value of each of the n calculation regions.

[0118] In operation S534, based on the objective function and the constraints, a quadratic programming problem with the air supply speed increment as a variable is constructed.

[0119] In operation S535, a numerically stable dual solution method is used to solve the quadratic programming problem to obtain an optimized air supply speed increment.

[0120] In operation S536, based on the optimized air supply speed increment, the optimized air supply speed at the current time k is determined so as to act on the air conditioning system to control the air supply speed when the supply and return air outlets in the target space transmit air.

[0121] For example, based on the state estimation model, it is possible to predict Temperature change of target space at each moment It can be shown as the following formula (21):

[0122] (twenty one)

[0123] l is the prediction time domain, The temperature of the target space at the moment It can be shown as the following formula (22):

[0124] (twenty two)

[0125] Among them, p=1, 2,…l.

[0126] The prediction time domain can be within predicted outputs (temperatures) and The input (air supply speed) is written in matrix form as shown in Equations (23) and (24):

[0127] (twenty three)

[0128] (twenty four)

[0129] The predicted temperature matrix of the target space in the prediction time domain can be obtained as shown in the following formula (25):

[0130] (25)

[0131] in, , , , .

[0132] The objective function can be constructed and the constraints of the air supply speed and the air supply speed increment can be set as shown in the following equation (26):

[0133] (26)

[0134] in, Represents the weight coefficient matrix, which is a diagonal matrix with dimension ; Represents the design temperature matrix of n calculation areas in the prediction time domain l, with dimensions of , the unit is ; They represent the minimum and maximum constraints in the air supply speed constraint column vector, respectively, and the dimension is , the unit is ; They represent the minimum and maximum constraints in the air supply speed increment constraint column vector, respectively, and the dimension is , the unit is .

[0135] Formula (25) can be substituted into formula (26), as shown in formula (27):

[0136] (27)

[0137] Formula (27) is organized into a quadratic programming form with the air supply speed increment as the control variable as shown in the following formula (28):

[0138] (28)

[0139] The constraints of the air supply speed and the increment of the air supply speed are sorted out to obtain the following formula (29):

[0140] (29)

[0141] in, represents the identity matrix, with dimensions ; represents a lower triangular matrix with elements equal to 1 and dimensions .

[0142] The quadratic programming problem can be rearranged as follows (30):

[0143] (30)

[0144] in, .

[0145] Since the objective function of formula (30) is a quadratic function and the constraints are linear constraints, it is a quadratic programming problem. We can use the numerically stable dual solution method to obtain the optimal solution that minimizes the function value. , take out the column vector The first element , determined by the following formula (31): .

[0146] (31)

[0147] in, represents the change in air supply speed at the current time k relative to the previous time k-1 of the current time k, represents the air supply speed at the previous moment k-1 before the current moment k, represents the air supply speed at the current time k. The air supply speed when the air supply and return outlets in the target space transmit air can be controlled as .

[0148] According to an embodiment of the present invention, through the energy balance equation, the air volume, air supply temperature, air density and other influencing factors of each calculation area are considered, combined with the linearization method, and the boundary condition differences of different calculation areas are considered, and the constructed state estimation model can reflect the dynamic characteristics of temperature changes in the actual target space environment. The target space air volume control problem based on state estimation is converted into a standard quadratic programming form, and a numerically stable dual solution method is used to solve the quadratic programming problem, and the variable air volume control of the target space is completed, which can quickly respond to fluctuations in environmental loads and enhance the control robustness of the system.

[0149] According to an embodiment of the present invention, by taking the air supply speed as the optimization variable and constructing a control problem with the goals of minimizing energy consumption and maximizing environmental comfort, it is possible to achieve global optimization of air volume distribution and energy utilization while ensuring the indoor environmental quality, thereby reducing the operating cost of the HVAC system.

[0150] According to the embodiments of the present invention, the variable air volume control method based on the Kalman filter proposed in the present invention is easy to integrate with the existing HVAC system, and has low dependence on hardware equipment. Only limited sensors need to be arranged in key areas to achieve efficient temperature field monitoring and optimization control. It can improve the operating performance of the HVAC system, reduce energy consumption, and help achieve green building and carbon neutrality goals.

[0151] Figure 6 A block diagram of a variable air volume control device based on a Kalman filter according to an embodiment of the present invention is shown.

[0152] like Figure 6 As shown, the variable air volume control device based on the Kalman filter includes a dividing module 610 , a processing module 620 , a determining module 630 and a controlling module 640 .

[0153] The division module 610 is used to divide the target space into n calculation areas based on a tetrahedral grid of a preset size. The processing module 620 is used to input the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k into the state estimation model when it is determined that the m temperature-measurable calculation areas meet the preset conditions, and output the real temperature increment matrix of the m temperature-measurable calculation areas. The determination module 630 is used to determine the posterior estimation matrix of the n calculation areas at the current time k based on the real temperature increment matrix of the m temperature-measurable calculation areas, the prior estimation matrix of the n calculation areas, and the gain matrix of the Kalman filter. The control module 640 is used to control the air supply speed when the air supply and return air outlets in the target space are transmitting air based on the constraints of the air supply speed and the air supply speed increment designed when the air supply and return air outlets transmit air, the posterior estimation matrix of the n calculation areas at the current time k, and the design temperature matrix of the n calculation areas.

[0154] According to an embodiment of the present invention, any multiple modules among the division module 610, the processing module 620, the determination module 630 and the control module 640 may be combined into one module for implementation, or any one of the modules may be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules may be combined with at least part of the functions of other modules and implemented in one module.

[0155] It should be noted that the variable air volume control device part based on the Kalman filter in the embodiment of the present invention corresponds to the variable air volume control method part based on the Kalman filter in the embodiment of the present invention. The description of the variable air volume control device part based on the Kalman filter specifically refers to the variable air volume control method part based on the Kalman filter, which will not be repeated here.

[0156] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention may be combined and / or combined in various ways. All of these combinations and / or combinations fall within the scope of the present invention.

[0157] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination advantageously. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A variable air volume control method based on Kalman filter, characterized in that: The variable air volume control method comprises: Based on a tetrahedral grid of a preset size, the target space is divided into n calculation areas, wherein the target space contains an air supply port and an air return port, the air supply port is used to transmit air from outside the target space to inside the target space, and the air return port is used to transmit air from inside the target space to outside the target space, each of the calculation areas includes at least one tetrahedral grid, n is a positive integer, and there are m calculation areas with measurable temperatures among the n calculation areas, m≤n, and m is a positive integer; When it is determined that the m temperature-measurable calculation areas meet the preset conditions, the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k are input into the state estimation model, and the real temperature increment matrix of the m temperature-measurable calculation areas is output, wherein the state estimation model is constructed based on a Kalman filter; Determine, based on the true temperature increment matrices of the m temperature-measurable calculation areas, the prior estimation matrices of the n calculation areas, and the gain matrix of the Kalman filter, the a posteriori estimation matrices of the n calculation areas at the current time k, wherein the gain matrix of the Kalman filter is determined based on the covariance matrix of the process noise and the covariance matrix of the measurement noise; and Based on the constraints on the designed air supply speed and air supply speed increment when the supply air outlet and the return air outlet transport air, the posterior estimation matrix of the n calculation areas at the current time k and the design temperature matrix of the n calculation areas, the air supply speed when the supply air outlet and the return air outlet transport air in the target space is controlled.

2. The variable air volume control method according to claim 1, characterized in that: The measured temperature increment matrix is ​​used to indicate the measured value of the temperature increment, the real temperature increment matrix is ​​used to indicate the real value of the temperature increment, and the temperature increment is used to indicate the temperature change at the current moment k relative to the previous moment k-1 at the current moment k; The variable air volume control method further comprises: Determine the covariance matrix of the posterior error according to the posterior error between the posterior estimated value of k at the current moment and the true value; Derivative the trace of the covariance matrix of the a posteriori error with respect to the gain matrix of the Kalman filter, and set the derivative to zero, to obtain an equation of the gain matrix of the Kalman filter with respect to the covariance matrix of the a priori error at the current time k and the covariance matrix of the measurement noise; Determine the covariance matrix of the a priori error at the current moment k based on the covariance matrix of the process noise and the covariance matrix of the a posteriori error at the previous moment k-1, wherein the covariance matrix of the a posteriori error at the previous moment k-1 is determined based on the a posteriori error of the a posteriori estimate value at the previous moment k-1, and the covariance matrix of the process noise is determined based on the temperature increments of the n calculation regions calculated by the state estimation model and the fluid dynamics model; and The gain matrix of the Kalman filter is determined based on the covariance matrix of the prior error at the current time k, the covariance matrix of the measurement noise and the equation, wherein the covariance matrix of the measurement noise is determined based on the measurement noise of the temperature sensor.

3. The variable air volume control method according to claim 2, characterized in that: The a priori estimation matrix is ​​used to indicate the a priori estimation value of the temperature increment, and the a priori estimation value is determined based on the a posteriori estimation value at the previous moment k-1; The variable air volume control method further comprises: Determining the temperature increments of the n calculation regions calculated by the state estimation model according to a priori estimation matrices of the n calculation regions; Taking the air supply speed when the air is transmitted by the air supply port and the air return port as a variable, using the fluid dynamics model for simulation calculation, to obtain the temperature increments of the n calculation areas calculated by the fluid dynamics model; determining a covariance matrix of the process noise according to the covariance of the temperature increments of the n calculation regions calculated by the state estimation model and the temperature increments of the n calculation regions calculated by the fluid dynamics model; and A diagonal matrix is ​​constructed according to the square of the measurement noise of the temperature sensor to obtain a covariance matrix of the measurement noise.

4. The variable air volume control method according to claim 3, characterized in that: The equation is shown in the following formula (1): （1） Among them, K(k) represents the gain matrix of the Kalman filter at the current time k, and its dimension is ; Represents the screening matrix of m calculation areas where the temperature can be measured, with dimension ; Represents the covariance matrix of the measurement noise, with dimension ; Represents the covariance matrix of the prior error at the current moment k, with dimension , n represents the number of the calculation regions, m represents the number of calculation regions with measurable temperatures, and T represents transposition.

5. The variable air volume control method according to claim 4, characterized in that: The covariance matrix of the prior error at the current time k is shown in the following formula (2): （2） in, Represents the covariance matrix of the posterior error of the previous moment k-1, with dimension ; Represents the covariance matrix of the process noise, with dimension , n represents the number of the calculation regions, m represents the number of the calculation regions with measurable temperatures, and T represents transposition; Represents the coefficient matrix of the temperature variable, with dimension .

6. The variable air volume control method according to claim 5, characterized in that: The variable air volume control method further comprises: The air flow in the target space is simulated by taking the air supply speed and the preset size when the air supply port or the return air port transmits air as variables to obtain a simulation result, wherein the simulation result is used to characterize the air supply volume of each mesh surface of the tetrahedral mesh at the preset size and the air supply speed; Based on the simulation results, determining a fitting relationship between the regional interface air volume and the air supply speed for the calculation area, wherein the regional interface air volume is used to characterize the sum of the air supply volumes of each grid surface in the calculation area; constructing an energy balance equation for each of the calculation regions based on the fitting relationship and the geometric position of each of the calculation regions in the target space; The energy balance equation is linearized and then converted into an incremental form to obtain an energy balance equation in an incremental form. According to the time step between the current time k and the previous time k-1, the energy balance equation in the incremental form is converted into a time discrete equation; and Based on the time-discrete equation, a coefficient matrix of the temperature variables is determined.

7. The variable air volume control method according to claim 1, characterized in that: The posterior estimation matrix of the n calculation regions at the current time k is determined as shown in the following formula (3): （3） in, represents the posterior estimation matrix of the n computational regions at the current time k, with a dimension of ; represents the prior estimation matrix of the n computational regions at the current time k, with a dimension of ; K(k) represents the gain matrix of the Kalman filter at the current time k, and its dimension is ; The real temperature increment matrix representing the m temperature measurable calculation areas has a dimension of ; Represents the screening matrix of m calculation areas where the temperature can be measured, with dimension .

8. The variable air volume control method according to any one of claims 1 to 7, characterized in that: The a posteriori estimation matrix is ​​used to indicate the a posteriori estimation value of the temperature increment; The control of the air supply speed when the air supply outlet and the return air outlet transmit air in the target space based on the constraints of the air supply speed and the air supply speed increment designed when the air supply outlet and the return air outlet transmit air, the posterior estimation matrix of the n calculation areas at the current time k, and the design temperature matrix of the n calculation areas, includes: Determine, based on the a posteriori estimation matrix of the n computational regions at the current time k, the respective a posteriori estimation values ​​of the n computational regions at the current time k; Determining the design temperature value of each of the n calculation regions based on the design temperature matrix of the n calculation regions; Determine the objective function according to minimizing the difference between the posterior estimation value and the design temperature value of each of the n calculation regions; Based on the objective function and the constraint conditions, construct a quadratic programming problem with the air supply speed increment as a variable; Solving the quadratic programming problem using a numerically stable dual solution method to obtain an optimized air supply speed increment; and Based on the optimized air supply speed increment, the optimized air supply speed at the current moment k is determined so that the air conditioning system controls the air supply speeds of the air supply outlet and the return air outlet in the target space according to the optimized air supply speed at the current moment k.

9. The variable air volume control method according to claim 8, characterized in that: The target space includes at least one of the following: an airport terminal, an exhibition center, an industrial plant, and a station.

10. A variable air volume control device based on Kalman filter, characterized in that: The variable air volume control device comprises: A division module, for dividing the target space into n calculation areas based on a tetrahedral grid of a preset size, wherein the target space contains an air supply port and an air return port, the air supply port is used to transmit air from outside the target space to inside the target space, and the air return port is used to transmit air from inside the target space to outside the target space, each of the calculation areas includes at least one tetrahedral grid, n is a positive integer, and there are m calculation areas with measurable temperatures among the n calculation areas, m≤n, and m is a positive integer; A processing module, for inputting the measured temperature noise matrix and the measured temperature increment matrix of the m temperature-measurable calculation areas at the current time k into a state estimation model when it is determined that the m temperature-measurable calculation areas meet a preset condition, and outputting the real temperature increment matrix of the m temperature-measurable calculation areas, wherein the state estimation model is constructed based on a Kalman filter; a determination module, configured to determine, based on the true temperature increment matrices of the m temperature-measurable calculation regions, the prior estimation matrices of the n calculation regions, and the gain matrix of the Kalman filter, the a posteriori estimation matrices of the n calculation regions at a current time k, wherein the gain matrix of the Kalman filter is determined based on the covariance matrix of the process noise and the covariance matrix of the measurement noise; and A control module is used to control the air supply speed when the air supply outlet and the return air outlet transport air in the target space based on the constraints of the designed air supply speed and air supply speed increment when the air supply outlet and the return air outlet transport air, the a posteriori estimation matrix of the n calculation areas at the current time k, and the design temperature matrix of the n calculation areas.

Images