Data center air conditioning equipment control method and system

Abstract

The invention discloses a data center air conditioning equipment control method and system. Comprising the steps of obtaining a first environment temperature and a second environment temperature of a target space where the data center air conditioning equipment is located at a first sampling moment and a second sampling moment, and determining a temperature error and an error change rate of the target space at the first sampling moment by combining an expected environment temperature, the temperature error and the error change rate serve as target input variables to be matched with a preset fuzzy rule base, a target output fuzzy set is obtained, defuzzification processing is conducted on the target output fuzzy set, and a corresponding target output variable is obtained; and a control signal is determined according to the temperature error, the error change rate and the target output variable, and equipment operation parameters of the data center air conditioning equipment are adjusted through the control signal. The technical problem that the effect of controlling the equipment operation parameters of the data center air conditioning equipment through the control signals generated through the fixed PID parameters is poor is solved.

Description

Technical Field

[0001] The present application relates to the technical field of equipment automatic control, and in particular to a method and system for controlling air-conditioning equipment in a data center. Background Art

[0002] In traditional computer room air conditioning control systems, PID (Proportional-Integral-Derivative) controllers are widely used to regulate ambient temperature, ensuring it remains within a set comfort range. The proportional coefficient Kp ensures a rapid response to temperature deviations, the integral coefficient Ki eliminates steady-state errors, and the differential coefficient Kd suppresses overshoot and overshoot.

[0003] However, traditional methods often rely on engineers' experience when determining PID control parameters. These parameters are fixed at the beginning of system design in order to achieve the desired control quality. In complex and changing computer room environments, these fixed PID control parameters may not provide optimal control and may even lead to deterioration in control performance, as shown in the following aspects:

[0004] (1) When the load suddenly increases, the fixed PID control parameters may not respond quickly, causing the room temperature to rise too quickly, and vice versa. This not only affects the stability of equipment operation, but may also lead to energy waste.

[0005] (2) Fixed PID control parameters cannot achieve optimal energy utilization efficiency under different load conditions, especially when the load is low, which may lead to unnecessary energy consumption.

[0006] (3) During the temperature control process, excessively large differential or integral terms may cause system oscillation and affect the long-term stable operation of the air-conditioning equipment.

[0007] Therefore, existing methods for setting PID parameters based on empirical values ​​are insufficient for complex and changing data center environments, resulting in technical issues such as poor control effectiveness, high energy consumption, and poor system stability. Therefore, there is an urgent need for a more intelligent and adaptive control strategy that can precisely adjust the room's ambient temperature while optimizing energy efficiency and system stability, providing strong support for efficient data center operations. Summary of the Invention

[0008] The embodiments of the present application provide a data center air conditioning equipment control method and system to at least solve the technical problem that the control signal generated by using fixed PID parameters is poorly effective in controlling the equipment operating parameters of the data center air conditioning equipment.

[0009] According to one aspect of an embodiment of the present application, a method for controlling data center air-conditioning equipment is provided, including: obtaining a first ambient temperature and a second ambient temperature of a target space where the data center air-conditioning equipment is located at a first sampling moment and a second sampling moment, wherein the second sampling moment is a sampling moment before the first sampling moment; determining a temperature error and an error change rate of the target space at the first sampling moment based on the first ambient temperature, the second ambient temperature and a preset expected ambient temperature, and matching the temperature error and the error change rate with a preset fuzzy rule base as target input variables to obtain a corresponding target output fuzzy set, defuzzifying the target output fuzzy set to obtain a corresponding target output variable, wherein the target output variable includes: a control parameter for generating a control signal and an adjustment parameter for adjusting the strength of the control signal, and the fuzzy rule base defines a mapping rule between multiple input fuzzy sets corresponding to the input variables and multiple output fuzzy sets corresponding to the output variables; determining a control signal based on the temperature error, the error change rate and the target output variable, and using the control signal to adjust the equipment operating parameters of the data center air-conditioning equipment.

[0010] Optionally, the construction process of the fuzzy rule base includes: determining input variables and output variables, wherein the input variables are temperature error and error change rate, and the control parameters in the output variables include: proportional parameter, differential parameter, integral parameter; determining multiple input fuzzy sets corresponding to different value ranges of the input variables and the first membership function corresponding to each input fuzzy set, and determining multiple output fuzzy sets corresponding to different value ranges of the output variables and the second membership function corresponding to each output fuzzy set, wherein the first membership function is used to describe the degree of membership between the input variable and each input fuzzy set, and the second membership function is used to describe the degree of membership between the output variable and each output fuzzy set; determining the mapping rules between each input fuzzy set and each output fuzzy set according to historical experience to obtain the fuzzy rule base.

[0011] Optionally, the temperature error and the error change rate are used as target input variables to match with a preset fuzzy rule base to obtain a corresponding target output fuzzy set, and the target output fuzzy set is defuzzified to obtain a corresponding target output variable, including: using the temperature error and the error change rate as target input variables, and using each first membership function to map the target input variable to each input fuzzy set to obtain a fuzzy vector corresponding to the target input variable, wherein the fuzzy vector includes: the membership degree of the target input variable to each input fuzzy set; matching the fuzzy vector with each mapping rule in the fuzzy rule base, and applying fuzzy logic operations to determine whether the successfully matched output fuzzy set is activated, wherein the fuzzy logic operations at least include: minimization operations or algebraic product operations; merging all successfully matched and activated output fuzzy sets to obtain a target output fuzzy set, and defuzzifying the target output fuzzy set to obtain a target output variable.

[0012] Optionally, all successfully matched and activated output fuzzy sets are merged to obtain a target output fuzzy set, including: obtaining the ambient temperature, control parameters and equipment operating parameters of the target space at each sampling moment in a first time window, wherein the first time window is a time window with the first sampling moment as the right endpoint; using a pre-trained temperature error prediction model to analyze the ambient temperature, control parameters and equipment operating parameters at each sampling moment in the first time window to obtain the predicted temperature error of the target space at each sampling moment in a second time window, wherein the second time window is a time window with the next sampling moment of the first sampling moment as the left endpoint; determining the activation intensity coefficient of each successfully matched and activated output fuzzy set based on the predicted ambient temperature error of the target space where the data center air-conditioning equipment is located at each sampling moment in the second time window; and performing weighted merging of each successfully matched and activated output fuzzy set based on the activation intensity coefficient to obtain the target output fuzzy set.

[0013] Optionally, the first ambient temperature includes at least one of the following: return air temperature and supply air temperature, wherein determining the control signal according to the temperature error, the error change rate, and the target output variable includes:

[0014]

[0015] Where K fuzzy (t0) represents the adjustment parameter at the first sampling time t0, K p (t0) represents the proportional parameter at the first sampling time t0, K i (t0) represents the integral parameter at the first sampling time t0, K d (t0) represents the differential parameter at the first sampling time t0, e(t0) represents the first ambient temperature error of the target space at the first sampling time t0, represents the error change rate at the first sampling time t0, u(t0) represents the control signal corresponding to the target space at the first sampling time t0, and the types of control signals include: a first control signal for controlling the fan speed of the data center air-conditioning equipment, and a second control signal for controlling the compressor frequency of the data center air-conditioning equipment.

[0016] Optionally, the equipment operating parameters of the data center air-conditioning equipment are adjusted using a control signal, including: when the first ambient temperature is the return air temperature, adjusting the fan speed of the data center air-conditioning equipment according to the first control signal; when the first ambient temperature is the supply air temperature, adjusting the compressor frequency of the data center air-conditioning equipment according to the second control signal.

[0017] Optionally, after adjusting the equipment operating parameters of the data center air conditioning equipment using the control signal, the method further includes: obtaining a third ambient temperature of the target space at each sampling moment in the second time window; determining a third ambient temperature difference of the target space at each sampling moment in the second time window based on each third ambient temperature; and calculating an equipment control effect of the data center air conditioning equipment in the second time window based on the following objective function:

[0018] J=w1·E+w2·O+w3·S

[0019] Where E represents the total energy consumption of the data center air conditioning equipment in the second time window, and t i Represents the time interval between each sampling moment in the second time window, Pin(t i ) represents the input power of the data center air conditioning equipment, Pin(t i )=Vf(t i ) 2 ·k1+Fc(t i ) 2 k2, where Vf(t i ) represents the i-th sampling time t of the data center air conditioning equipment in the second time window i The fan speed under i ) represents the i-th sampling time t of the data center air conditioning equipment in the second time window i The compressor frequency under the condition of , k1, k2 are the fan characteristic parameters and compressor characteristic parameters respectively; O represents the maximum absolute value of the ambient temperature error of the data center air conditioning equipment in the second time window, and Among them, T set Indicates the expected ambient temperature, Represents the target space at the i-th sampling time t in the second time window i The third ambient temperature under the third time window; S represents the root mean square of the ambient temperature error of the data center air conditioning equipment in the second time window, and Where N represents the number of sampling moments in the second time window, The target space is at the i-th sampling time t in the second time window i The third ambient temperature difference under the control effect threshold is obtained; w1, w2, and w3 represent weight coefficients, and w1+w2+w3=1; when the equipment control effect is lower than the preset control effect threshold, the particle swarm optimization algorithm is used to iteratively optimize the target output variable to obtain the optimized target output variable whose equipment control effect is not lower than the control effect threshold; a new control signal is determined based on the temperature error, the error change rate, and the optimized target output variable, and the equipment operating parameters of the data center air-conditioning equipment are readjusted using the new control signal.

[0020] According to another aspect of the embodiment of the present application, a data center air-conditioning equipment control system is also provided, including: an acquisition module, used to obtain the first ambient temperature and the second ambient temperature of the target space where the data center air-conditioning equipment is located at the first sampling moment and the second sampling moment, wherein the second sampling moment is the sampling moment before the first sampling moment; a matching module, used to determine the temperature error and the error change rate of the target space at the first sampling moment based on the first ambient temperature, the second ambient temperature and the preset expected ambient temperature, and match the temperature error and the error change rate as target input variables with a preset fuzzy rule base to obtain a corresponding target output fuzzy set, defuzzify the target output fuzzy set to obtain a corresponding target output variable, wherein the target output variable includes: a control parameter for generating a control signal and an adjustment parameter for adjusting the strength of the control signal, and the fuzzy rule base defines a mapping rule between multiple input fuzzy sets corresponding to the input variables and multiple output fuzzy sets corresponding to the output variables; a control module, used to determine the control signal based on the temperature error, the error change rate and the target output variable, and use the control signal to adjust the equipment operating parameters of the data center air-conditioning equipment.

[0021] According to another aspect of an embodiment of the present application, a computer program product is further provided, the computer program product comprising: a computer program, wherein when the computer program is executed by a processor, the above-mentioned data center air conditioning equipment control method is implemented.

[0022] According to another aspect of an embodiment of the present application, an electronic device is provided, which includes: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-mentioned data center air conditioning equipment control method through the computer program.

[0023] In an embodiment of the present application, the control system obtains the ambient temperature of the target space where the data center air conditioning equipment is located at two consecutive sampling moments in real time, and calculates the temperature error and error change rate of the target space at the current sampling moment based on these temperature data and the preset expected ambient temperature; the calculated temperature error and error change rate are used as target input variables, matched with the preset fuzzy rule base, and the target output fuzzy set is obtained through fuzzy logic reasoning; the target output fuzzy set is then defuzzified, and a clear control signal is generated in combination with the temperature error. Therefore, the embodiment of the present application adjusts the air outlet process of the data center air conditioning equipment according to the obtained control signal by integrating the flexibility of fuzzy control and the accuracy of PID control, ensuring that the ambient temperature in the target space is maintained at the expected ambient temperature, thereby achieving the technical effects of dynamically adapting to environmental changes, optimizing control effects, and saving energy and reducing consumption. This solves the technical problem of poor effect of controlling the equipment operating parameters of the data center air conditioning equipment using a control signal generated by fixed PID parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0025] Figure 1 is a flow chart of an optional data center air conditioning equipment control method according to an embodiment of the present application;

[0026] Figure 2 is a schematic structural diagram of an optional data center air conditioning equipment control system according to an embodiment of the present application;

[0027] Figure 3 It is a schematic structural diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0029] It should be noted that the terms "first", "second", etc. in the specification, claims, and drawings of the present application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0030] In order to better understand the embodiments of the present application, some nouns or terms that appear in the description of the embodiments of the present application are first translated and explained as follows:

[0031] PID (Proportion Integration Differentiation) control is an automatic control measurement widely used in industrial process control. It generates a control signal by performing proportional, integral, and differential operations on the system's deviation, thereby achieving precise control of the controlled object. Therefore, the core of the PID control system is divided into the following three parts: (1) Proportional control: Proportional control generates a control signal based on the current deviation value. The larger the proportional coefficient, the faster the system response, but it may cause system instability; (2) Integral control: Integral control generates a control signal based on the accumulated value of the deviation, mainly used to eliminate the steady-state error of the system; (3) Differential control: Differential control generates a control signal based on the rate of change of the deviation, mainly used to predict the trend of the deviation change and improve the dynamic performance of the system.

[0032] Fuzzy set: A set used to express fuzzy concepts, also known as fuzzy set or fuzzy subset. Specifically, given a domain U, a mapping μ from U to the unit interval [0,1] is A :U→[0,1] is called a fuzzy set on U, or a fuzzy subset of U. Where, the fuzzy set and as A, the mapping function μ A (·) or simply A(·) is called the membership function of the fuzzy set A.

[0033] Membership function: A statistical tool used to characterize fuzzy sets. To describe the membership of an element u to a fuzzy set on U, due to the ambiguity of this relationship, a numerical value from the interval [0, 1] is used instead of 0 and 1 to represent the "real degree" of the element's belonging to the fuzzy set.

[0034] Particle Swarm Optimization (PSO): also known as particle swarm optimization or bird flock foraging algorithm, this algorithm is a type of evolutionary algorithm. It is based on the observation of animal cluster activity behavior and uses the information sharing of individuals in the group to make the movement of the entire group evolve from disorder to order in the problem-solving space, thereby obtaining the optimal solution.

[0035] Example 1

[0036] According to an embodiment of the present application, a method for controlling air-conditioning equipment in a data center is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0037] Figure 1FIG. 1 is a flow chart of a method for controlling a data center air conditioning device according to an embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0038] Step S102 : obtaining a first ambient temperature and a second ambient temperature of a target space where the air-conditioning equipment of the data center is located at a first sampling moment and a second sampling moment.

[0039] In the technical solution provided in the above step S102, the control system obtains temperature monitoring using temperature sensors installed at key positions in the target space where the data center air conditioning equipment is located, wherein the temperature sensor can continuously monitor and record the ambient temperature according to a preset sampling period (such as 5 minutes), thereby providing a continuous temperature data stream. Therefore, the first sampling moment and the second sampling moment in the embodiment of the present application are two consecutive sampling moments, and the second sampling moment is the sampling moment before the first sampling moment. In addition, the collected ambient temperature can be the supply air temperature and the return air temperature measured by the temperature sensor, which respectively reflect the temperature of the air sent into the target space by the data center air conditioning equipment and the temperature of the air recovered from the target space.

[0040] Step S104: determine the temperature error and error change rate of the target space at the first sampling moment based on the first ambient temperature, the second ambient temperature and the preset expected ambient temperature, and use the temperature error and error change rate as target input variables to match them with the preset fuzzy rule library to obtain the corresponding target output fuzzy set, defuzzify the target output fuzzy set, and obtain the corresponding target output variable.

[0041] In the technical solution provided in step S104 above, the control system calculates the difference (i.e., temperature error) between the first ambient temperature and the set temperature at the current moment (i.e., the first sampling moment) and the degree of change in this error relative to the previous moment (i.e., the second sampling moment) (i.e., the error rate of change). The calculated temperature error and error rate of change are used as inputs to the fuzzy control system and matched against a pre-set fuzzy rule base. The fuzzy rule base defines mapping relationships between multiple input fuzzy sets corresponding to input variables (i.e., temperature error and error rate of change) and multiple output fuzzy sets corresponding to output variables (i.e., control parameters for generating control signals and adjustment parameters for adjusting the strength of control signals). These mapping rules are pre-set based on expert knowledge and system behavior characteristics. Through fuzzy reasoning, the system can find the target output fuzzy set that best matches the current state. Subsequently, the system defuzzifies the target output fuzzy set to convert the result of fuzzy logic reasoning (i.e., the target output fuzzy set) into clear control parameters and adjustment parameters (i.e., target output variables) to adjust the control signals for the supply and return air temperatures, thereby achieving precise temperature regulation of the target space.

[0042] Step S106 : determining a control signal based on the temperature error and the target output variable, and adjusting the equipment operating parameters of the data center air conditioning equipment using the control signal.

[0043] In the technical solution provided in step S106 above, the control system applies the target output variable determined by fuzzy inference to the PID control equation and combines it with the measured temperature error to calculate a corresponding control signal. This control signal instructs the data center air conditioning equipment how to adjust its outlet air temperature and air volume to achieve the desired ambient temperature.

[0044] Based on the scheme defined by the above steps S102 to S106, it can be known that in the embodiment of the present application, the control system obtains the ambient temperature of the target space where the data center air conditioning equipment is located at two consecutive sampling moments in real time, and calculates the temperature error and error change rate of the target space at the current sampling moment based on these temperature data and the preset expected ambient temperature; the calculated temperature error and error change rate are used as target input variables, matched with the preset fuzzy rule base, and the target output fuzzy set is obtained through fuzzy logic reasoning; the target output fuzzy set is then defuzzified, and a clear control signal is generated in combination with the temperature error. Therefore, the embodiment of the present application adjusts the air outlet process of the data center air conditioning equipment according to the obtained control signal by integrating the flexibility of fuzzy control and the accuracy of PID control, ensuring that the ambient temperature in the target space is maintained at the expected ambient temperature, thereby achieving the technical effects of dynamically adapting to environmental changes, optimizing control effects, and saving energy and reducing consumption.

[0045] The following describes the steps of the data center air conditioning equipment control method in combination with a specific implementation process.

[0046] In the technical solution provided in the above step S104, the control system can first calculate the temperature error of the target space at the first sampling moment and the second temperature error of the target space at the second sampling moment based on the first ambient temperature, the second ambient temperature and the preset expected ambient temperature; then calculate the difference between the second temperature error and the temperature error to obtain the error change rate of the target space at the first sampling moment.

[0047] For example, assuming that the expected ambient temperature T of the target space at the first sampling moment is setis 24°C, and the first ambient temperature T(t) of the target space at the first sampling moment is 23°C, and the first ambient temperature T(t-1) of the target space at the first sampling moment is 25°C, then the temperature error e(t) of the target space at the first sampling moment is 1°C, and the second temperature error e(t-1) of the target space at the second sampling moment is -1°C, and the error change rate of the target space at the first sampling moment is e(t)-e(t-1)=1-(-1)=2°C.

[0048] As an optional implementation, in the technical solution provided in step S104 above, the control system may construct a fuzzy rule base according to the following steps, including:

[0049] Step S1, determine the input variables and output variables. The input variables are the temperature error e(t) and the error change rate Δe(t), and the control parameters in the output variables include: the proportional parameter K p , differential parameter K i , integration parameter K d , and the adjustment parameter in the output variable can be recorded as K fuzzy .

[0050] Step S2: determining a plurality of input fuzzy sets corresponding to different value ranges of the input variables and a first membership function corresponding to each input fuzzy set, and determining a plurality of output fuzzy sets corresponding to different value ranges of the output variables and a second membership function corresponding to each output fuzzy set.

[0051] Specifically, multiple fuzzy sets are defined for the input variables, each representing a fuzzy state of error or rate of change, such as "extremely negative," "large negative," and "small negative," to describe different levels of temperature error and rate of change. Next, a first membership function is selected for each input fuzzy set to quantify the degree of membership between the input variable and each input fuzzy set. Similarly, multiple fuzzy sets are defined for the output variables, such as "low," "medium," and "high," to describe the fuzzy level of the control parameter. A second membership function is selected for each output fuzzy set to quantify the degree of membership between the output variable and each output fuzzy set. Common types of first and second membership functions include triangular membership functions and Gaussian membership functions.

[0052] In step S3, mapping rules between input fuzzy sets and output fuzzy sets are determined based on historical experience to obtain a fuzzy rule base. These mapping rules reflect how the PID control system parameters should be adjusted to achieve the best control effect under different temperature errors and error change rates.

[0053] Therefore, the control system can call the above-mentioned fuzzy rule library to intelligently output control parameters and adjustment parameters suitable for the current temperature state based on the input temperature error and error change rate, thereby achieving precise control of the data center air-conditioning equipment.

[0054] The construction of the fuzzy rule base will be explained in detail below.

[0055] First, for each input and output variable, a membership function is defined to describe its fuzzy set. A Gaussian membership function is typically used, with a standard deviation of 0.5. This is because, as verified by Monte Carlo simulations, a standard deviation of 0.5 achieves the optimal balance between fuzzy inference response speed (average response time 0.8 seconds) and stability (error fluctuation ±0.3°C).

[0056] For example, the membership function of the temperature error e(t) within the input variable uses a Gaussian distribution, and the fuzzy levels of its fuzzy set can be defined as: extremely negative (NVB), negative large (NB), negative small (NS), zero (Z), positive small (PS), positive large (PB), and extremely positive (PVB). Therefore, the membership function table for the temperature error e(t) can be referred to in Table 1 below.

[0057] Table 1

[0058] Fuzzy level Parameter range (℃) Center value (μ) Standard deviation (σ) Very large negative (NVB) [-5,-3] -4 0.5 Negative Big (NB) [-4,-2] -3 0.5 Negative Small (NS) [-2,0] -1 0.5 Zero (Z) [-0.5,0.5] 0 0.2 Small (PS) [0,2] 1 0.5 CP Group (PB) [2,4] 3 0.5 PVB [3,5] 4 0.5

[0059] The membership function for the temperature error change rate Δe(t) within the input variable uses a triangular distribution, and the fuzzy levels of its fuzzy set can be defined as: extremely negative (NVB), negative large (NB), negative small (NS), zero (Z), positive small (PS), positive large (PB), and extremely positive (PVB). Therefore, the membership function table for the temperature error change rate Δe(t) can be found in Table 2 below.

[0060] Table 2

[0061] Fuzzy level Left border (a) Vertex (b) Right border (c) Very large negative (NVB) -0.8 -0.6 -0.4 Negative Big (NB) -0.6 -0.4 -0.2 Negative Small (NS) -0.3 -0.1 0 Zero (Z) -0.1 0 0.1 Small (PS) 0 0.1 0.3 CP Group (PB) 0.2 0.4 0.6 PVB 0.4 0.6 0.8

[0062] In addition, the proportional parameter K in the output variable p , differential parameter K i , integration parameter K d and tuning parameter K fuzzy The membership functions of all adopt triangular distribution, and the fuzzy levels of the fuzzy sets can be defined as: low (L), medium (M), high (H), and very high (VH). fuzzy The value range of can be [0.5, 1.5]. Therefore, the membership function table of each variable in the output variable can refer to the following Table 3.

[0063] Table 3

[0064]

[0065] Furthermore, the following rules are formulated through traditional PID parameter tuning using the expert experience method (white box model):

[0066] (1)K p rule:

[0067] When the absolute value of the temperature error e(t) is large (such as the fuzzy level of the temperature error is NVB / PVB) and the temperature error change trend is intensified (i.e. the temperature error e(t) and the temperature error change rate Δe(t) have the same sign), the control system can adopt high K p value (VH / H) so that the control system can respond quickly to the current temperature deviation.

[0068] When the absolute value of the temperature error e(t) is medium (e.g., the fuzzy level of the temperature error is NS / PS) or the temperature error changes slowly (i.e., the temperature error e(t) and the temperature error change rate Δe(t) have different signs), the control system can use a medium K p value (M), which can ensure that the system has sufficient response capability and avoid instability or oscillation caused by over-regulation.

[0069] When the temperature error e(t) is close to zero (Z) or the temperature error change tends to be stable (i.e. the temperature error change rate Δe(t) tends to zero), the control system can use low K p value (L) to prevent unnecessary over-regulation. This is because low K p The value (L) can reduce the control system's sensitivity to small changes and avoid making large adjustments when the temperature is very close to the set expected ambient temperature, which may cause temperature overshoot or temperature fluctuation.

[0070] (2)K i rule:

[0071] When the system has a stable temperature error for a long time, that is, the temperature error e(t) is continuously non-zero and the temperature error change rate Δe(t) approaches zero, the control system can use a high K i value (VH / H), so that the control system accumulates past errors more strongly so that the system can continue to adjust until the residual error is completely eliminated and accurate tracking of the set temperature is achieved.

[0072] In the case where the system is actively adjusting in response to temperature deviations, that is, the temperature error e(t) and the rate of change of the temperature error Δe(t) have the same sign, the control system can use a medium K iThis is because: moderate integral action can ensure that the system gradually eliminates errors in the dynamic process, and at the same time, by timely resetting the integral term or limiting its growth rate, it avoids the occurrence of integral saturation problems, allowing the system to respond more flexibly to future temperature changes.

[0073] When the temperature error change rate Δe(t) changes greatly, it means that the system is in a transition period when the load suddenly increases or decreases. In order to avoid excessive accumulation of integral action in such cases, which may cause excessive system response or oscillation, the control system can use a low K i Value (L). This is because: Low K i The value reduces the influence of the integral term on the control system output, allowing the system to respond more quickly to current errors without being affected by the accumulation of past errors, thereby improving the temperature control accuracy and stability under dynamic load conditions.

[0074] (3)K d rule:

[0075] When the temperature error e(t) is changing rapidly, that is, the temperature error change rate Δe(t) is large (such as the fuzzy level of the temperature error change rate is NVB / PVB), the control system can use high K d This is because a high Kd value can increase the control system's sensitivity to the error change rate, allowing the control system to respond more quickly to error changes and suppress excessive temperature oscillations in a timely manner.

[0076] When the rate of change of the temperature error e(t) is moderate, that is, the temperature error change rate Δe(t) is a medium value (such as the fuzzy level of the temperature error change rate is NS / PS), the control system can adopt the medium K d value (M). This is because, d The value neither relies too much on the fast response of the differential action nor ignores the importance of the differential action for smooth control, so as to help the system avoid temperature fluctuations caused by overly drastic adjustments while maintaining a certain response speed, thereby improving the overall smoothness of the control process.

[0077] When the system is close to a stable state, that is, the temperature error change rate Δe(t) is very small (that is, Δe(t) approaches zero), the control system can use a low K d This is because, in steady state or when changes are extremely slow, the system's differential effect on temperature control becomes more sensitive, easily introducing noise interference and leading to unnecessary control actions. A low Kd value can reduce the impact of the differential term on the overall control signal, avoiding overreaction of the system due to tiny measurement noise, thereby maintaining smooth system operation, reducing energy consumption, and improving overall control quality.

[0078] Therefore, based on the above rule principles, the proportional parameter K can be formulated separately p , differential parameter K i , integration parameter K d The respective fuzzy rule tables are shown in Table 4-6 below.

[0079] Table 4

[0080]

[0081] Table 5

[0082]

[0083]

[0084] Table 6

[0085]

[0086] In addition, the above adjustment parameter K fuzzy The rules are as follows:

[0087] When the absolute value of the temperature error e(t) is large (such as the fuzzy level of the temperature error is NVB / PVB) and the temperature error change trend is intensified (i.e. the temperature error e(t) and the temperature error change rate Δe(t) have the same sign), the control system can use high K p Value (such as blur level is VH / H) and high K fuzzy Value (such as the fuzzy level is VH / H). Among them, high K fuzzy The value can further strengthen the control signal, ensuring that the system can respond quickly and quickly reduce the temperature difference.

[0088] When the system has a stable temperature error for a long time, that is, the temperature error e(t) is continuously non-zero and the temperature error change rate Δe(t) approaches zero, the control system can use a high K d Value (H) and medium K fuzzy value (such as the fuzzy level is M). Among them, K fuzzy The value can ensure that the control signal is not too strong, prevent the temperature from being over-adjusted during the process of eliminating the residual error, and keep the control process stable.

[0089] When the temperature error e(t) is changing rapidly, that is, the temperature error change rate Δe(t) is large (such as the fuzzy level of the temperature error change rate is NVB / PVB), the control system can use high K d Value (H) and high K fuzzy Value. Among them, high K fuzzy The value can also serve to strengthen the control signal.

[0090] Furthermore, the system can use the temperature error and the error change rate as target input variables to match the fuzzy rule base constructed above in the following steps to obtain the corresponding target output fuzzy set, and defuzzify the target output fuzzy set to obtain the corresponding target output variable, including:

[0091] Step S1041 : taking the temperature error and the error change rate as target input variables, and using each first membership function to map the target input variables to each input fuzzy set, to obtain a fuzzy vector corresponding to the target input variable.

[0092] In the technical solution provided in step S1041 above, for each input fuzzy set, the system uses the first membership function of each input fuzzy set to calculate the degree to which the target input variable (i.e., temperature error and error rate of change) belongs to that fuzzy set (i.e., membership degree), and represents this degree as a value between 0 and 1. For example, if the temperature error falls within the "negative large" fuzzy set, then the corresponding membership degree of the "negative large" fuzzy set will be higher; if the error rate of change is close to zero, then the membership degree of the "zero" fuzzy set will be higher. Subsequently, the degree to which the target input variable belongs to each fuzzy set is summarized to obtain the corresponding fuzzy vector.

[0093] Step S1042 : Match the fuzzy vector with each mapping rule in the fuzzy rule base, and apply fuzzy logic operation to determine whether the output fuzzy set of the successful match is activated.

[0094] In the technical solution provided in step S1042 above, the system matches the generated fuzzy vector with each rule in the fuzzy rule base and uses fuzzy logic operations to determine which rules should be activated. Common fuzzy logic operations include at least minimization operations or algebraic product operations. These operations can help determine which output fuzzy sets should be triggered when the target input variable falls into a specific fuzzy set.

[0095] Step S1043 : All successfully matched and activated output fuzzy sets are merged to obtain a target output fuzzy set, and the target output fuzzy set is defuzzified to obtain a target output variable.

[0096] In the technical solution provided in the above step S1043, after all rule matching is completed, all activated output fuzzy sets are merged (i.e., intersection or union is taken) to obtain a target output fuzzy set; the target output fuzzy set is then defuzzified. There are various defuzzification methods, among which the most common ones are the centroid defuzzifier method and the maximum membership method. These methods can calculate a specific numerical value to convert the result obtained by fuzzy logic reasoning into an actual executable control parameter for output.

[0097] Because the control system needs to combine the activation strength coefficients of the output fuzzy sets when merging all successfully matched and activated output fuzzy sets. The relevant technology usually determines the activation strength coefficients of each successfully matched and activated output fuzzy set based on the degree of matching between the input fuzzy variables and the fuzzy sets described in the rules. For example, if the membership degree of the temperature error e(t) belonging to "extremely negative" is 0.8, and the membership degree of the error change rate Δe(t) belonging to "extremely negative" is also 0.8, then the activation strength of the output fuzzy set in the matching mapping rule is the product of these two memberships, that is, 0.64 (when algebraic product merging is used). Since this static determination method relies on preset fuzzy rules and membership functions, it is difficult to self-adjust according to real-time changing environmental conditions or control effects, which limits the performance of the controller in a dynamic environment.

[0098] To this end, an embodiment of the present application proposes to dynamically adjust the activation intensity coefficient of each successfully matched and activated output fuzzy set by introducing an online learning algorithm, so as to merge all successfully matched and activated output fuzzy sets to obtain a target output fuzzy set.

[0099] Optionally, in the technical solution provided in step S1043 above, the system may merge all successfully matched and activated output fuzzy sets according to the following method, including:

[0100] Step 1: Obtain the ambient temperature, control parameters, and equipment operating parameters of the target space at each sampling moment within the first time window, wherein the first time window is a time window with the first sampling moment as the right endpoint.

[0101] Step 2: Use the pre-trained temperature error prediction model to analyze the ambient temperature, control parameters, and equipment operating parameters at each sampling time within the first time window to obtain the predicted temperature error of the target space at each sampling time within the second time window. The temperature error prediction model can be a recurrent neural network, and the second time window is the time window with the sampling time immediately following the first sampling time as its left endpoint.

[0102] Step 3: Determine the activation strength coefficients of each successfully matched and activated output fuzzy set based on the predicted temperature error at each sampling moment within the second time window for the target space where the data center air conditioning equipment is located. If the predicted temperature error is large, and the trend of the predicted temperature error is consistent with the control direction of the output fuzzy set, then the activation strength coefficient of the corresponding output fuzzy set is likely to be higher, indicating that its control strategy is more effective in responding to future trends. Conversely, if the predicted temperature error is small, or the trend of the predicted temperature error is opposite to the control direction of the output fuzzy set, then the activation strength coefficient of the corresponding output fuzzy set is likely to be lower, indicating that its control effect may be weak in the future, and its weight in the fuzzy control output needs to be appropriately reduced.

[0103] Step 4: Based on the activation strength coefficient, each successfully matched and activated output fuzzy set is weightedly merged to obtain the target output fuzzy set. This weighted merging can employ various fuzzy logic operations, such as weighted averaging or maximum membership in triangular fuzzy logic, to ensure that the outputs of each fuzzy set are combined according to their activation strength. As a result, the merged target output fuzzy set reflects the comprehensive fuzzy signal of future temperature control requirements.

[0104] In the above embodiment, the activation intensity coefficients for each successfully matched and activated output fuzzy set are no longer determined statically, but are instead based on the ambient temperature error predicted by the online learning algorithm. This means that the activation intensity coefficients can be dynamically adjusted based on current and future environmental changes, making control more flexible and precise. Furthermore, this determination method reduces reliance on expert knowledge, easing the difficulty of system debugging and maintenance, and improving the intelligence of the control system.

[0105] For example, assuming that the expected ambient temperature T of the target space at the first sampling moment is set is 24°C, the first ambient temperature T(t) of the target space at the first sampling moment is 26°C, and the temperature error of the target space at the second sampling moment is e rf (t-1) = -1°C. Then, the temperature error e of the target space at the first sampling moment is rf (t) = T set -T rf =24-26=-2℃, the error change rate of the target space at the first sampling moment is: Δe rf (t) = e rf (t)-e rf (t-1) = -1°C.

[0106] According to Table 1 above, e rf(t) = -2℃ belongs to "Negative Large (NB)" with a membership of 0.5, and belongs to "Negative Small (NS)" with a membership of 0.5. rf (t) = -1°C has a membership degree of 0.5 belonging to "Negative Small (NS)" and a membership degree of 0.5 belonging to "Zero (Z)".

[0107] Then, fuzzy reasoning is performed through Table 4-6. Among them, after consulting the scale parameter K shown in Table 4 above, p From the fuzzy rule table, we can see that if e rf (t) is NB and Δe rf (t) is NS, then K p p is M; if e rf (t) is NB and Δe rf (t) is Z, then K p is L; if e rf (t) is NS and Δe rf (t) is NS, then K p is M; if e rf (t) is NS and Δe rf (t)f is Z, then K p It’s L.

[0108] Therefore, if we assume that L corresponds to K p The activation intensity coefficient of the value is 0.3, M corresponds to K p The activation intensity coefficient of the value is 0.5, then the scale parameter K p The value of is: K p =(0.5*0.5+0.5*0.3+0.5*0.5+0.5*0.3) / (0.5+0.5+0.5+0.5)=0.4.

[0109] As an optional implementation, in the technical solution provided in step S106 above, the control system can combine fuzzy PID control parameters with real-time monitoring values ​​of environmental parameters to form a closed-loop control model to dynamically adjust the equipment operating parameters of the data center air conditioning equipment. Therefore, the control system can apply the defuzzified target output parameters to the following PID control formula to calculate the control signal as follows:

[0110]

[0111] Where K fuzzy (t0) represents the adjustment parameter at the first sampling time t0, K p (t0) represents the proportional parameter at the first sampling time t0, K i (t0) represents the integral parameter at the first sampling time t0, K d(t0) represents the differential parameter at the first sampling time t0, e(t0) represents the first ambient temperature error of the target space at the first sampling time t0, represents the error change rate at the first sampling time t0, and u(t0) represents the control signal corresponding to the target space at the first sampling time t0.

[0112] For example, if the desired ambient temperature T set The first ambient temperature T(t) of the target space at the first sampling moment is 25°C, the first ambient temperature T(t) of the target space at the first sampling moment is 24°C, the first ambient temperature T(t-1) of the target space at the first sampling moment is 23.5°C, the time interval between the first sampling moment and the second sampling moment is Δt=1 minute, and the control parameter is: K fuzzy (t0)=1,K p (t0)=1,K i (t0=0.1,K d (t0)=0.5.

[0113] Then, the temperature error e(t) of the target space at the first sampling moment is 1°C, and the second temperature error e(t-1) of the target space at the second sampling moment is 1.5°C. Then the integral of the error ∫e(t0)dt≈1*1=1, and the rate of change of the error is Then the final control signal can be expressed as: u(t0)=1·[1·1+0.1·1+0.5·(-0.5)]=0.85.

[0114] Since the first ambient temperature includes at least one of the following: return air temperature e rf (t0), air supply temperature e sf (t0), therefore, the types of control signals include: a first control signal for controlling the fan speed of the data center air conditioning equipment, and a second control signal for controlling the compressor frequency of the data center air conditioning equipment.

[0115] When the first ambient temperature is the return air temperature, the expression of the corresponding first control signal can be written as:

[0116]

[0117] Therefore, the control system can adjust the fan speed of the data center air conditioning equipment according to the first control signal Vf(t0).

[0118] When the first ambient temperature is the supply air temperature, the expression of the corresponding second control signal can be written as:

[0119]

[0120] Therefore, the control system can adjust the compressor frequency of the data center air conditioning equipment according to the second control signal Fc(t0).

[0121] As an optional implementation, after the control system uses the control signal to adjust the equipment operating parameters of the data center air conditioning equipment, it can also evaluate and optimize the equipment control effect through the following steps, including:

[0122] Step 1: Obtain the third ambient temperature of the target space at each sampling moment within the second time window.

[0123] Step 2: Determine the third ambient temperature difference of the target space at each sampling time within the second time window based on each third ambient temperature, and calculate the equipment control effect of the data center air conditioning equipment within the second time window based on the following objective function:

[0124] J=w1·E+w2·O+w3·S

[0125] Where E represents the total energy consumption of the data center air conditioning equipment in the second time window, and Δt i Represents the time interval between each sampling moment in the second time window, Pin(t i ) represents the input power of the data center air conditioning equipment, Pin(t i )=Vf(t i ) 2 ·k1+Fc(t i ) 2 k2, where Vf(t i ) represents the i-th sampling time t of the data center air conditioning equipment in the second time window i The fan speed under i ) represents the i-th sampling time t of the data center air conditioning equipment in the second time window i The compressor frequency under , k1, k2 are fan characteristic parameters and compressor characteristic parameters respectively; O represents the maximum absolute value of the ambient temperature error of the data center air conditioning equipment in the second time window (used to quantify the deviation of temperature control in the second time window), and Among them, T set Indicates the expected ambient temperature, Represents the target space at the i-th sampling time t in the second time window i S represents the root mean square of the ambient temperature error of the data center air conditioning equipment in the second time window (used to quantify the stability of temperature control in the second time window), and Where N represents the number of sampling moments in the second time window, The target space is at the i-th sampling time t in the second time window i The third ambient temperature difference under ; w1, w2, w3 represent weight coefficients, and w1+w2+w3=1.

[0126] Step 3: When the device control effect falls below the preset control effect threshold, the particle swarm optimization algorithm is used to iteratively optimize the target output variable, obtaining an optimized target output variable that ensures the device control effect is no less than the control effect threshold. When using the particle swarm optimization algorithm for optimization, the target output parameter can be used as a particle to search for the optimal output variable combination that minimizes J. The population size parameter, number of iterations, and learning factor can all be customized based on the actual application scenario.

[0127] Step 4: Determine a new control signal based on the temperature error, error change rate, and optimized target output variable, and use the new control signal to readjust the equipment operating parameters of the data center air conditioning equipment.

[0128] The evaluation and optimization process, through real-time data collection, multi-metric performance evaluation, adaptive parameter optimization, and control signal updates, ensured the data center's air conditioning equipment could continue to operate efficiently and stably in a dynamic environment, while also achieving the goals of minimizing energy consumption and maximizing temperature control accuracy. Specifically, through intelligent optimization using the particle swarm optimization algorithm, the control system was able to dynamically adjust control parameters to respond to load changes and ambient temperature fluctuations, thereby further improving energy efficiency and the intelligence level of the control system while meeting control requirements.

[0129] In the data center air conditioning equipment control method introduced above, a fuzzy rule base based on expert experience and Monte Carlo simulation is designed to determine the target output fuzzy set that matches the current temperature state of the target space where the data center air conditioning equipment is located, and the activation intensity coefficient of the target output fuzzy set is determined by predicting the ambient temperature difference at multiple sampling moments in the future based on the online learning algorithm, so that the system can predict environmental changes to a certain extent and respond in advance, avoiding the problem of control lag and enhancing the stability and accuracy of temperature control. By defuzzifying the target output fuzzy set, the corresponding target output variable (i.e., control parameters and adjustment parameters) is obtained, and finally the control signal is determined based on the temperature error, error change rate and target output variable, and the control signal is used to adjust the equipment operating parameters of the data center air conditioning equipment. In addition, the embodiment of the present application also uses a feedback mechanism and a particle swarm optimization algorithm to evaluate and optimize the generated control signal, which greatly improves the performance of the control system.

[0130] Example 2

[0131] According to an embodiment of the present application, a data center air conditioning equipment control system is also provided for implementing the data center air conditioning equipment control method in Example 1, such as Figure 2 As shown, the data center air conditioning equipment control system includes at least: an acquisition module 22, a matching module 24 and a control module 26, wherein:

[0132] The acquisition module 22 is configured to acquire the first ambient temperature and the second ambient temperature of the target space where the air-conditioning equipment of the data center is located at a first sampling moment and a second sampling moment, wherein the second sampling moment is a sampling moment before the first sampling moment.

[0133] The matching module 24 is used to determine the temperature error and error change rate of the target space at the first sampling moment based on the first ambient temperature, the second ambient temperature and the preset expected ambient temperature, and match the temperature error and error change rate as target input variables with a preset fuzzy rule base to obtain a corresponding target output fuzzy set, and defuzzify the target output fuzzy set to obtain a corresponding target output variable, wherein the target output variable includes: a control parameter for generating a control signal and an adjustment parameter for adjusting the strength of the control signal, and the fuzzy rule base defines mapping rules between multiple input fuzzy sets corresponding to the input variables and multiple output fuzzy sets corresponding to the output variables.

[0134] The control module 26 is used to determine a control signal based on the temperature error, the error change rate and the target output variable, and use the control signal to adjust the equipment operating parameters of the data center air conditioning equipment.

[0135] The following describes the functions of each module of the data center air conditioning equipment control system in combination with the specific implementation process.

[0136] Specifically, the matching module 24 can first calculate the temperature error of the target space at the first sampling moment and the second temperature error of the target space at the second sampling moment based on the first ambient temperature, the second ambient temperature and the preset expected ambient temperature; then calculate the difference between the second temperature error and the temperature error to obtain the error change rate of the target space at the first sampling moment.

[0137] As an optional implementation, the data center air conditioning equipment control system further includes a construction module 28, and the construction module 28 is used to construct a fuzzy rule base according to the following steps, including:

[0138] Step S1, determine the input variables and output variables. The input variables are the temperature error e(t) and the error change rate Δe(t), and the control parameters in the output variables include: the proportional parameter K p , differential parameter K i , integration parameter K d , and the adjustment parameter in the output variable can be recorded as Kfuzzy .

[0139] Step S2: determining a plurality of input fuzzy sets corresponding to different value ranges of the input variables and a first membership function corresponding to each input fuzzy set, and determining a plurality of output fuzzy sets corresponding to different value ranges of the output variables and a second membership function corresponding to each output fuzzy set.

[0140] Specifically, multiple fuzzy sets are defined for the input variables, each representing a fuzzy state of error or rate of change, such as "extremely negative," "large negative," and "small negative," to describe different levels of temperature error and rate of change. Next, a first membership function is selected for each input fuzzy set to quantify the degree of membership between the input variable and each input fuzzy set. Similarly, multiple fuzzy sets are defined for the output variables, such as "low," "medium," and "high," to describe the fuzzy level of the control parameter. A second membership function is selected for each output fuzzy set to quantify the degree of membership between the output variable and each output fuzzy set. Common types of first and second membership functions include triangular membership functions and Gaussian membership functions.

[0141] In step S3, mapping rules between input fuzzy sets and output fuzzy sets are determined based on historical experience to obtain a fuzzy rule base. These mapping rules reflect how the PID control system parameters should be adjusted to achieve the best control effect under different temperature errors and error change rates.

[0142] Therefore, the matching module 24 can call the above-mentioned fuzzy rule library to intelligently output control parameters and adjustment parameters suitable for the current temperature state based on the input temperature error and error change rate, thereby achieving precise control of the data center air-conditioning equipment.

[0143] Optionally, the matching module 24 may obtain the target output variable according to the following steps, including:

[0144] In step S1 , the temperature error and the error change rate are used as target input variables, and the target input variables are mapped to the input fuzzy sets using the first membership functions to obtain the fuzzy vectors corresponding to the target input variables.

[0145] That is, for each input fuzzy set, matching module 24 uses the first membership function of each input fuzzy set to calculate the degree to which the target input variable (i.e., temperature error and error rate of change) belongs to that fuzzy set (i.e., membership degree), and represents this degree as a value between 0 and 1. For example, if the temperature error falls within the "negative large" fuzzy set, then the corresponding membership degree of the "negative large" fuzzy set will be higher; if the error rate of change is close to zero, then the membership degree of the "zero" fuzzy set will be higher. Matching module 24 then summarizes the degree to which the target input variable belongs to each fuzzy set to obtain the corresponding fuzzy vector.

[0146] Step S2: Match the fuzzy vector with each mapping rule in the fuzzy rule base, and apply fuzzy logic operation to determine whether the output fuzzy set of the successful match is activated.

[0147] That is, the matching module 24 matches the generated fuzzy vector with each rule in the fuzzy rule base and uses fuzzy logic operations to determine which rules should be activated. Common fuzzy logic operations include at least minimization operations or algebraic product operations, which can help determine which output fuzzy sets should be triggered when the target input variable falls into a specific fuzzy set.

[0148] Step S3: merge all successfully matched and activated output fuzzy sets to obtain a target output fuzzy set, and perform defuzzification processing on the target output fuzzy set to obtain a target output variable.

[0149] That is to say, after all rule matching is completed, the matching module 24 can dynamically adjust the activation intensity coefficient of each successfully matched and activated output fuzzy set by introducing an online learning algorithm to merge all successfully matched and activated output fuzzy sets to obtain a target output fuzzy set; then the target output fuzzy set is defuzzified. There are many defuzzification methods, among which the most common ones are the centroid defuzzifier method, the maximum membership method, etc. Through these methods, a specific numerical value can be calculated to convert the result obtained by fuzzy logic reasoning into an actual executable control parameter for output.

[0150] Optionally, in the technical solution provided in the above step S3, the matching module 24 may merge all successfully matched and activated output fuzzy sets by the following method, including:

[0151] Step 1: Obtain the ambient temperature, control parameters, and equipment operating parameters of the target space at each sampling moment within the first time window, wherein the first time window is a time window with the first sampling moment as the right endpoint.

[0152] Step 2: Use the pre-trained temperature error prediction model to analyze the ambient temperature, control parameters, and equipment operating parameters at each sampling time within the first time window to obtain the predicted temperature error of the target space at each sampling time within the second time window. The temperature error prediction model can be a recurrent neural network, and the second time window is the time window with the sampling time immediately following the first sampling time as its left endpoint.

[0153] Step 3: Determine the activation strength coefficients of each successfully matched and activated output fuzzy set based on the predicted temperature error at each sampling moment within the second time window for the target space where the data center air conditioning equipment is located. If the predicted temperature error is large, and the trend of the predicted temperature error is consistent with the control direction of the output fuzzy set, then the activation strength coefficient of the corresponding output fuzzy set is likely to be higher, indicating that its control strategy is more effective in responding to future trends. Conversely, if the predicted temperature error is small, or the trend of the predicted temperature error is opposite to the control direction of the output fuzzy set, then the activation strength coefficient of the corresponding output fuzzy set is likely to be lower, indicating that its control effect may be weak in the future, and its weight in the fuzzy control output needs to be appropriately reduced.

[0154] Step 4: Based on the activation strength coefficient, each successfully matched and activated output fuzzy set is weightedly merged to obtain the target output fuzzy set. This weighted merging can employ various fuzzy logic operations, such as weighted averaging or maximum membership in triangular fuzzy logic, to ensure that the outputs of each fuzzy set are combined according to their activation strength. As a result, the merged target output fuzzy set reflects the comprehensive fuzzy signal of future temperature control requirements.

[0155] As an optional implementation, the control module 26 can combine fuzzy PID control parameters with real-time monitoring values ​​of environmental parameters to form a closed-loop control model to dynamically adjust the equipment operating parameters of the data center air conditioning equipment. Therefore, the control system can apply the defuzzified target output parameters to the following PID control formula to calculate the control signal as follows:

[0156]

[0157] Where K fuzzy (t0) represents the adjustment parameter at the first sampling time t0, K p (t0) represents the proportional parameter at the first sampling time t0, K i (t0) represents the integral parameter at the first sampling time t0, K d (t0) represents the differential parameter at the first sampling time t0, e(t0) represents the first ambient temperature error of the target space at the first sampling time t0, represents the error change rate at the first sampling time t0, and u(t0) represents the control signal corresponding to the target space at the first sampling time t0.

[0158] Since the first ambient temperature includes at least one of the following: return air temperature e rf (t0), air supply temperature e sf (t0), therefore, the types of control signals include: a first control signal for controlling the fan speed of the data center air conditioning equipment, and a second control signal for controlling the compressor frequency of the data center air conditioning equipment.

[0159] When the first ambient temperature is the return air temperature, the expression of the corresponding first control signal can be written as:

[0160]

[0161] Therefore, the control module 26 can adjust the fan speed of the data center air conditioning equipment according to the first control signal Vf(t0).

[0162] When the first ambient temperature is the supply air temperature, the expression of the corresponding second control signal can be written as:

[0163]

[0164] Therefore, the control module 26 can adjust the compressor frequency of the data center air conditioning equipment according to the second control signal Fc(t0).

[0165] It should be noted that each module in the data center air conditioning equipment control system in the embodiment of the present application corresponds one-to-one to each implementation step of the data center air conditioning equipment control method in Example 1. Since a detailed description has been given in Example 1, some details not reflected in this embodiment can be referred to Example 1 and will not be elaborated here.

[0166] Example 3

[0167] According to an embodiment of the present application, a computer program product is further provided. The computer program product includes a computer program, wherein when the computer program is executed by a processor, the data center air conditioning equipment control method in Example 1 is implemented.

[0168] According to an embodiment of the present application, a non-volatile storage medium is also provided, which includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the data center air conditioning equipment control method in Example 1 by running the computer program.

[0169] According to an embodiment of the present application, a processor is further provided, which is used to run a computer program, wherein the data center air conditioning equipment control method in Example 1 is executed when the computer program is run.

[0170] According to an embodiment of the present application, an electronic device is also provided, which includes: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the data center air conditioning equipment control method in Example 1 through the computer program.

[0171] Specifically, when the computer program is running, the following steps are executed: obtaining the first ambient temperature and the second ambient temperature of the target space where the data center air-conditioning equipment is located at the first sampling moment and the second sampling moment, wherein the second sampling moment is the sampling moment before the first sampling moment; determining the temperature error and the error change rate of the target space at the first sampling moment based on the first ambient temperature, the second ambient temperature and the preset expected ambient temperature, and matching the temperature error and the error change rate with the preset fuzzy rule base as target input variables to obtain the corresponding target output fuzzy set, defuzzifying the target output fuzzy set to obtain the corresponding target output variable, wherein the target output variable includes: control parameters for generating control signals and adjustment parameters for adjusting the strength of the control signal, and the fuzzy rule base defines mapping rules between multiple input fuzzy sets corresponding to the input variables and multiple output fuzzy sets corresponding to the output variables; determining the control signal based on the temperature error, the error change rate and the target output variable, and using the control signal to adjust the equipment operating parameters of the data center air-conditioning equipment.

[0172] As an optional implementation, the electronic device may be in the form of a mobile terminal, a computer terminal or a similar computing device. Figure 3 The hardware structure block diagram of an electronic device for implementing a data center air conditioning equipment control method is shown. Figure 3 As shown, the electronic device 30 may include one or more (illustrated as 302a, 302b, ..., 302n in the figure) processors 302 (the processor 302 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 304 for storing data, and a transmission device 306 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It will be understood by those skilled in the art that Figure 3 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 3 More or fewer components than shown, or with Figure 3 Different configurations shown.

[0173] It should be noted that the one or more processors 302 and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the electronic device 30. As described in the embodiments of the present application, the data processing circuitry serves as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0174] The memory 304 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the data center air conditioning equipment control method in the embodiment of the present application. The processor 302 executes various functional applications and data processing by running the software programs and modules stored in the memory 304, that is, implementing the vulnerability detection method of the above-mentioned application. The memory 304 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 304 may further include a memory remotely located relative to the processor 302, and these remote memories may be connected to the electronic device 30 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0175] The transmission device 306 is used to receive or send data via a network. Specific examples of the aforementioned network may include a wireless network provided by the communications provider of the electronic device 30. In one embodiment, the transmission device 306 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In one embodiment, the transmission device 306 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0176] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the electronic device 30 .

[0177] The serial numbers of the above embodiments are for description only and do not represent the advantages or disadvantages of the embodiments.

[0178] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0179] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0180] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected to achieve the purpose of the present embodiment according to actual needs.

[0181] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0182] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program code.

[0183] The above is only a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A data center air conditioning equipment control method, characterized in that: include: Obtaining a first ambient temperature and a second ambient temperature of a target space where the air conditioning equipment of the data center is located at a first sampling time and a second sampling time, wherein the second sampling time is a sampling time before the first sampling time; Determining a temperature error and an error change rate of the target space at a first sampling moment based on the first ambient temperature, the second ambient temperature, and a preset expected ambient temperature, matching the temperature error and the error change rate as target input variables with a preset fuzzy rule base to obtain a corresponding target output fuzzy set, performing defuzzification processing on the target output fuzzy set to obtain a corresponding target output variable, wherein the target output variable includes: a control parameter for generating a control signal and an adjustment parameter for adjusting the strength of the control signal, and the fuzzy rule base defines mapping rules between multiple input fuzzy sets corresponding to the input variables and multiple output fuzzy sets corresponding to the output variables; A control signal is determined according to the temperature error, the error change rate, and the target output variable, and the equipment operating parameters of the data center air conditioning equipment are adjusted using the control signal.

2. The method according to claim 1, characterized in that The construction process of the fuzzy rule base includes: Determine input variables and output variables, wherein the input variables are temperature error and error change rate, and the control parameters in the output variables include: proportional parameter, differential parameter, and integral parameter; Determining a plurality of input fuzzy sets corresponding to different value ranges of the input variable and a first membership function corresponding to each of the input fuzzy sets, and determining a plurality of output fuzzy sets corresponding to different value ranges of the output variable and a second membership function corresponding to each of the output fuzzy sets, wherein the first membership function is used to describe the degree of membership between the input variable and each of the input fuzzy sets, and the second membership function is used to describe the degree of membership between the output variable and each of the output fuzzy sets; The mapping rules between the input fuzzy sets and the output fuzzy sets are determined according to historical experience to obtain the fuzzy rule base.

3. The method according to claim 2, characterized in that The temperature error and the error change rate are used as target input variables to match a preset fuzzy rule base to obtain a corresponding target output fuzzy set, and the target output fuzzy set is defuzzified to obtain a corresponding target output variable, including: The temperature error and the error change rate are used as target input variables, and the target input variables are mapped to the input fuzzy sets using the first membership functions to obtain fuzzy vectors corresponding to the target input variables, wherein the fuzzy vectors include: the membership degrees of the target input variables to the input fuzzy sets; Matching the fuzzy vector with each mapping rule in the fuzzy rule base, and applying a fuzzy logic operation to determine whether a successfully matched output fuzzy set is activated, wherein the fuzzy logic operation at least includes: a minimization operation or an algebraic product operation; All successfully matched and activated output fuzzy sets are merged to obtain the target output fuzzy set, and the target output fuzzy set is defuzzified to obtain the target output variable.

4. The method according to claim 3, characterized in that All successfully matched and activated output fuzzy sets are merged to obtain the target output fuzzy set, including: Obtaining the ambient temperature, control parameters, and equipment operating parameters of the target space at each sampling time within a first time window, wherein the first time window has the first sampling time as a right endpoint; Analyze the ambient temperature, control parameters, and equipment operating parameters at each sampling time within the first time window using a pre-trained temperature error prediction model to obtain a predicted temperature error of the target space at each sampling time within a second time window, where the second time window is a time window with the next sampling time after the first sampling time as its left endpoint; Determining the activation intensity coefficient of each successfully matched and activated output fuzzy set according to the predicted ambient temperature error of the target space where the data center air conditioning equipment is located at each sampling time within the second time window; According to the activation intensity coefficient, each successfully matched and activated output fuzzy set is weightedly combined to obtain the target output fuzzy set.

5. The method according to claim 1, characterized in that The first ambient temperature includes at least one of the following: return air temperature and supply air temperature. The control signal is determined according to the temperature error, the error change rate, and the target output variable, including: Where K fuzzy (t0) represents the adjustment parameter at the first sampling time t0, K p (t0) represents the proportional parameter at the first sampling time t0, K i (t0) represents the integral parameter at the first sampling time t0, K d (t0) represents the differential parameter at the first sampling time t0, e(t0) represents the first ambient temperature error of the target space at the first sampling time t0, represents the error change rate at the first sampling time t0, u(t0) represents the control signal corresponding to the target space at the first sampling time t0, and the types of the control signal include: a first control signal for controlling the fan speed of the data center air-conditioning equipment, and a second control signal for controlling the compressor frequency of the data center air-conditioning equipment.

6. The method according to claim 5, characterized in that Using the control signal to adjust the equipment operating parameters of the data center air conditioning equipment includes: When the first ambient temperature is the return air temperature, adjusting the fan speed of the data center air conditioning equipment according to the first control signal; When the first ambient temperature is the supply air temperature, the compressor frequency of the data center air conditioning equipment is adjusted according to the second control signal.

7. The method according to claim 1, characterized in that After adjusting the equipment operating parameters of the data center air conditioning equipment using the control signal, the method further includes: Acquire a third ambient temperature of the target space at each sampling moment within the second time window; Determine the third ambient temperature difference of the target space at each sampling moment in the second time window based on each of the third ambient temperatures, and calculate the equipment control effect of the data center air conditioning equipment in the second time window based on the following objective function: J=w1·E+w2·O+w3·S Where E represents the total energy consumption of the data center air conditioning equipment in the second time window, and t i Represents the time interval between each sampling moment in the second time window, Pin(t i ) represents the input power of the data center air conditioning equipment, Pin(t i )=Vf(t i ) 2 ·k1+Fc(t i ) 2 k2, where Vf(t i ) represents the i-th sampling time t of the data center air conditioning equipment in the second time window i The fan speed under i ) represents the i-th sampling time t of the data center air conditioning equipment in the second time window i The compressor frequency under the condition of , k1, k2 are fan characteristic parameters and compressor characteristic parameters respectively; O represents the maximum absolute value of the ambient temperature error of the data center air conditioning equipment in the second time window, and Among them, T set represents the desired ambient temperature, Indicates the target space at the i-th sampling time t in the second time window i S represents the root mean square of the ambient temperature error of the data center air conditioning equipment in the second time window, and Wherein, N represents the number of sampling moments in the second time window, The target space is at the i-th sampling time t in the second time window i The third ambient temperature difference under the condition of φ1; w1, w2, w3 represent weight coefficients, and w1+w2+w3=1; When the device control effect is lower than a preset control effect threshold, the target output variable is iteratively optimized using a particle swarm optimization algorithm to obtain an optimized target output variable whose device control effect is not lower than the control effect threshold; A new control signal is determined based on the temperature error, the error change rate, and the optimized target output variable, and the equipment operating parameters of the data center air conditioning equipment are readjusted using the new control signal.

8. A data center air conditioning equipment control system, characterized in that: include: An acquisition module, configured to acquire a first ambient temperature and a second ambient temperature of a target space where the air-conditioning equipment of the data center is located at a first sampling moment and a second sampling moment, wherein the second sampling moment is a sampling moment before the first sampling moment; a matching module, configured to determine a temperature error and an error change rate of the target space at a first sampling moment based on the first ambient temperature, the second ambient temperature, and a preset expected ambient temperature, and to match the temperature error and the error change rate as target input variables with a preset fuzzy rule base to obtain a corresponding target output fuzzy set, and to perform defuzzification processing on the target output fuzzy set to obtain a corresponding target output variable, wherein the target output variable includes: a control parameter for generating a control signal and an adjustment parameter for adjusting the strength of the control signal, and the fuzzy rule base defines mapping rules between multiple input fuzzy sets corresponding to input variables and multiple output fuzzy sets corresponding to output variables; A control module is used to determine a control signal based on the temperature error, the error change rate and the target output variable, and adjust the equipment operating parameters of the data center air conditioning equipment using the control signal.

9. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the data center air conditioning equipment control method according to any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the data center air conditioning equipment control method according to any one of claims 1 to 7 through the computer program.

Images