A high-robust central air conditioning energy-saving control method

Abstract

The present application relates to the technical field of air conditioning energy-saving control, and discloses a high-robust central air conditioning energy-saving control method, which comprises the following steps: constructing a subway central air conditioning robust energy-saving regulation and control model to formally represent the adaptive temperature control and regulation process of the subway central air conditioning system; constructing a robust energy-saving regulation and control objective function according to the subway central air conditioning robust energy-saving regulation and control model; collecting current short-term energy consumption time series data of the subway central air conditioning, optimizing and solving the robust energy-saving regulation and control objective function, and obtaining the energy-saving regulation and control strategy of the subway central air conditioning at the next moment. The present application realizes energy consumption prediction of various energy-consuming devices, solves the regulation and control parameters of the energy-consuming devices in the subway central air conditioning based on the energy consumption at the next moment obtained through prediction and the air conditioning supply air temperature, and further regulates and controls the energy-consuming devices in the subway central air conditioning, so as to realize adaptive parameter adjustment and adaptive temperature adjustment of the subway central air conditioning based on short-term energy consumption.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of air conditioning energy-saving control, and particularly relates to a high-robust central air conditioning energy-saving control method. BACKGROUND

[0002] With the rapid development of the subway industry, the power consumption also increases year by year. The subway air conditioning system needs to run uninterruptedly throughout the year. At present, the power consumption of the subway air conditioning system in China is about 202.8 billion kilowatt-hours, accounting for about 2.7% of the total power consumption in China, and the energy consumption cannot be ignored. Due to the large number of devices and complex loops of the subway air conditioning system, which contains nonlinear devices such as fans and water pumps, it shows strong nonlinear, strong coupling and large lag characteristics. The traditional PID control parameter setting is difficult, and it is difficult to achieve good control effect, and it is difficult to realize the overall energy efficiency improvement of the system. In view of this problem, the present application provides a high-robust central air conditioning energy-saving control method to realize intelligent energy-saving control of the subway air conditioning. SUMMARY

[0003] Therefore, the present application provides a high-robust central air conditioning energy-saving control method, which aims to: 1) based on the short-term energy consumption time series data of the energy consumption devices of the subway central air conditioning, using the subway central air conditioning energy consumption time series prediction model to predict the energy consumption of different energy consumption devices at the next moment, converting the predicted energy consumption into air conditioning supply air temperature, realizing the mapping of short-term energy consumption time series data and air conditioning supply air temperature, and then taking the minimum energy consumption as the target, taking the air conditioning supply air temperature range as the constraint, taking the temperature difference between the regulated temperature and the predicted temperature as the reward value to construct the strategy regulation function, and solving to obtain the energy-saving regulation strategy that satisfies the constraint condition and has the minimum temperature difference, wherein the smaller the temperature difference is, the closer the regulated temperature is to the predicted temperature, and the predicted result is similar to the time series characteristics of the short-term energy consumption time series data, the predicted result is smoother, and the difference between the regulated temperature and the previous temperature is smaller, thereby avoiding large changes in air conditioning supply air temperature; 2) using two groups of parameter vector solutions to solve the parameters to be solved in the subway central air conditioning energy consumption time series prediction model, in the generation process of the initial solution, a generation strategy combining random weights is used to generate more extensive and more random initial solutions, expanding the initial space of the parameter vector solution, in the early stage of algorithm iteration, an exploratory solution updating scheme combining random solutions is used to expand the search range of the parameter vector solution, and the model parameters are quickly solved, and in the later stage of algorithm iteration, the parameter vector solution is updated in an antagonistic competition manner, if the antagonistic difference between the two is large, the parameter updating range is expanded, the local optimal solution is avoided, and the effectiveness of the solved parameters is ensured.

[0004] To achieve the above-mentioned purposes, the present application provides a high-robust central air conditioning energy-saving control method, which comprises the following steps:

[0005] S1: Construct a time series prediction model for the energy consumption of the central air conditioning system in the subway. The model takes short-term energy consumption time series data as input and the energy consumption at the next moment as output.

[0006] S2: Construct a temperature simulation model for the central air conditioning system in the subway. The model takes energy consumption as input and air conditioning supply temperature as output.

[0007] S3: Construct a robust energy-saving control model for subway central air conditioning. The constructed robust energy-saving control model for subway central air conditioning includes a time-series prediction model for subway central air conditioning energy consumption, a temperature simulation model for subway central air conditioning, and an air conditioning energy consumption control model. The adaptive temperature control process of the subway central air conditioning system is formally represented.

[0008] S4: Construct a robust energy-saving control objective function based on the robust energy-saving control model of subway central air conditioning;

[0009] S5: Collect the current short-term energy consumption time series data of the subway central air conditioning, optimize and solve the constructed robust energy-saving control objective function, and obtain the energy-saving control strategy of the subway central air conditioning in the next moment.

[0010] As a further improvement of the present invention:

[0011] Optionally, the construction of the time-series prediction model for the energy consumption of the subway central air conditioning system in step S1 includes:

[0012] A time-series prediction model for the energy consumption of subway central air conditioning is constructed. This model takes short-term energy consumption time-series data as input and outputs the energy consumption at the next time step. The short-term energy consumption time-series data is in the following format:

[0013]

[0014] in:

[0015] Indicates t h Energy consumption of the central air conditioning system in the subway at all times. P c (t h ) represents t h

[0016] The cooling load provided by the chiller unit in the central air conditioning system of the subway at all times, P e (t h ) represents t h The operating power (P) of the cooling tower fan in the central air conditioning system of the subway at any given time. s (t h ) represents t h The operating power of the water pumps in the central air conditioning system of the subway at all times;

[0017] Indicates time t1 to t H Short-term energy consumption time-series data at any given moment will As the input to the time-series prediction model for energy consumption of subway central air conditioning, the model prediction output is t. H+1 Energy consumption of subway central air conditioning

[0018] Based on short-term energy consumption time series data As input, the energy consumption prediction process based on the time-series prediction model for subway central air conditioning is as follows:

[0019] Set the maximum number of prediction iterations to d. num :

[0020]

[0021] in:

[0022] num represents the number of moments in the short-term energy consumption time series data, and th1 represents the preset iteration threshold.

[0023] d min This represents the expected number of prediction iterations set in advance;

[0024] Set the current prediction iteration number to d, with an initial value of 1;

[0025] Repeat the following prediction iteration formula until the d-th term is obtained. num The result of the dth iteration, and the dth iteration result. num The result of the next iteration is used as the predicted t. H+1 Energy consumption of subway central air conditioning

[0026]

[0027] L 3,d-1 =σ[w3(L 1,d-1 +L 2,d-1 )+b3]

[0028]

[0029]

[0030] in:

[0031] σ(·) represents the activation function. In this embodiment of the invention,

[0032] w1, w2, w3, w4 represent weight parameters, and b1, b2, b3, b4 represent bias parameters;

[0033] P d This represents the result of the d-th iteration. This is the iteration matrix.

[0034] Optionally, the optimization of parameters in the time-series prediction model for the energy consumption of the subway central air conditioning system in step S1 includes:

[0035] The parameters to be optimized in the time-series prediction model for the energy consumption of the subway central air conditioning system include weight parameters w1, w2, w3, w4 and bias parameters b1, b2, b3, b4. These parameters are constructed as a parameter vector [w1, w2, w3, w4, b1, b2, b3, b4]. The parameter optimization process is as follows:

[0036] S11: Training dataset for constructing a time-series prediction model for the energy consumption of subway central air conditioning:

[0037] data={(Y k ,y k |k∈[1,K]}

[0038] in:

[0039] Y k This represents the k-th training data set in the training dataset `data`. Each training data set consists of short-term energy consumption time-series data of varying lengths. k Y represents k Actual energy consumption data at the next moment;

[0040] S12: Constructing the training objective function F(β) for the time-series prediction model of energy consumption of subway central air conditioning:

[0041]

[0042] in:

[0043] β represents the parameter vector to be optimized in the time series prediction model of energy consumption of subway central air conditioning;

[0044] Indicate Y k The input is fed into the β-based time series prediction model for energy consumption of subway central air conditioning, and the model outputs the predicted value.

[0045] S13: Set the current number of iterations for parameter optimization to m, and the maximum number of iterations to Max. The initial value of m is 1. Generate 2N parameter vector solutions, and divide the generated parameter vector solutions into two groups, each group containing N parameter vector solutions. The formula for generating the parameter vector solutions is:

[0046]

[0047] in:

[0048] rand(0,1) represents a random number between 0 and 1, β max This represents the upper boundary of the pre-defined parameter vector to be optimized, β. min This represents the lower boundary of the pre-defined parameter vector to be optimized;

[0049] Let i represent the nth parameter vector solution in the i-th group, where i = 1, 2, n ∈ [1, N];

[0050] The iteration result of the nth parameter vector solution in the i-th group after the m-th iteration is: Solution of parameter vector The fitness value of the parameter vector solution is obtained by inputting it into the training objective function.

[0051] S14: Construct the training process judgment threshold Temp:

[0052]

[0053] S15: If Temp < 0.25, it indicates that the current stage is exploration, and the two sets of parameter vector solutions are iteratively updated; otherwise, proceed to step S16, where the iterative update formula is:

[0054]

[0055] in:

[0056] Represents the solution of the random parameter vector of the i-th group;

[0057] This represents the iteration result of the nth parameter vector solution in the i-th group after the m-th iteration;

[0058] Let m = m + 1, then return to step S15;

[0059] S16: If Temp ≥ 0.25, it indicates that the current stage is competitive, and the two sets of parameter vector solutions are iteratively updated, where the iterative update formula is:

[0060]

[0061]

[0062]

[0063]

[0064] in:

[0065] It represents the parameter vector solution with the minimum fitness value in the first group as of the current iteration number. It represents the parameter vector solution with the minimum fitness value in the second group as of the current iteration number.

[0066] If m < Max, then m = m + 1, and return to step S16. Otherwise, calculate the fitness values of all current parameter vector solutions, select the parameter vector solution with the minimum fitness value as the optimized parameter in the subway central air-conditioning energy consumption time series prediction model, and construct the subway central air-conditioning energy consumption time series prediction model based on the optimized parameter.

[0067] Optionally, in step S2, constructing the subway central air-conditioning temperature simulation model includes:

[0068] Constructing the subway central air-conditioning temperature simulation model based on the SVR algorithm. The constructed subway central air-conditioning temperature simulation model takes energy consumption as the input and the air supply temperature of the air conditioner as the output, that is, establishing the hyperplane mapping relationship between energy consumption and the air supply temperature of the air conditioner:

[0069] M(P) = wP + b

[0070] Where:

[0071] wP + b represents the constructed hyperplane, P represents energy consumption, M(P) represents the air supply temperature corresponding to P, w represents the linear weight of the hyperplane wP + b, and b represents the bias value of the hyperplane wP + b;

[0072] Obtain training data, and establish a mean square error loss function based on the training data. Use the particle swarm optimization algorithm to optimize and solve the hyperplane parameters w and b, and construct the hyperplane mapping relationship between energy consumption and the air supply temperature of the air conditioner in the subway central air-conditioning temperature simulation model based on the obtained hyperplane parameters. In the embodiment of the present invention, the form of the training data of the subway central air-conditioning temperature simulation model is (energy consumption, air supply temperature), and the acquisition process of the training data is to collect the energy consumption of the subway central air-conditioning at different times and the corresponding air supply temperature at the corresponding time to form multiple groups of training data.

[0073] Optionally, in step S3, constructing the subway central air-conditioning robust energy-saving regulation model includes:

[0074] Constructing the subway central air-conditioning robust energy-saving regulation model. The constructed subway central air-conditioning robust energy-saving regulation model includes the subway central air-conditioning energy consumption time series prediction model, the subway central air-conditioning temperature simulation model, and the air-conditioning energy consumption regulation model. The air-conditioning energy consumption regulation model is used to adjust the parameters of the subway central air-conditioning, thereby regulating the energy consumption of the subway central air-conditioning. The regulated energy consumption includes the cooling load provided by the chiller in the subway central air-conditioning, the operating power of the cooling tower fan, and the operating power of the water pump. The corresponding energy consumption regulation formula is:

[0075] P c =cq co (W1-W2)

[0076]

[0077]

[0078] in:

[0079] c represents the specific heat capacity of water;

[0080] q co This indicates the chilled water flow rate supplied by the chiller unit, W1 represents the chilled water supply temperature, W2 represents the chilled water return temperature, and P... c q represents the cooling load provided by the chiller units in the central air conditioning system of the subway. co W1 and W2 represent the control parameters of the chiller unit;

[0081] f1 represents the operating frequency of the cooling tower fan, and f0 represents the rated frequency of the cooling tower fan. P represents the rated power of the cooling tower fan. e This indicates the operating power of the cooling tower fan in the subway central air conditioning system, and f1 represents the control parameters of the cooling tower fan.

[0082] ρ represents the density of water, g represents the acceleration due to gravity, and H represents the acceleration due to gravity. s This indicates the head of the water pump in the subway's central air conditioning system;

[0083] V represents the flow rate of the water pump, ρ s ρ represents the efficiency of a water pump. m P represents the efficiency of the frequency converter. s V,ρ represents the operating power of the water pump in the subway central air conditioning system. s ,ρ m These represent the control parameters of the water pump;

[0084] By adjusting the energy-consuming equipment parameters in the subway central air conditioning system in real time, the energy consumption after adjustment is obtained. The energy-consuming equipment includes chillers, cooling tower fans, and water pumps.

[0085] Optionally, in step S3, the adaptive temperature control process of the subway central air conditioning system is formally represented based on the robust energy-saving control model of the subway central air conditioning system, including:

[0086] Based on the robust energy-saving control model of subway central air conditioning, the adaptive temperature control process of the subway central air conditioning system is formally represented. The formally represented adaptive temperature control process is as follows:

[0087] S31: Collect the current short-term energy consumption time series data of the subway central air conditioning system, and input the collected results into the subway central air conditioning energy consumption time series prediction model to obtain the energy consumption at the next moment.

[0088] S32: Input the energy consumption at the next moment into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation results;

[0089] S33: Based on the predicted energy consumption at the next moment and the corresponding simulation results of the subway central air conditioning temperature, the control parameters of the energy-consuming equipment in the subway central air conditioning are solved based on the high robust energy-saving control target, and the energy-consuming equipment in the subway central air conditioning is controlled based on the solved control parameters, so as to realize the adaptive parameter adjustment and adaptive temperature adjustment of the subway central air conditioning based on short-term energy consumption.

[0090] Optionally, the construction of the robust energy-saving control objective function in step S4 includes:

[0091] Construct a robust energy-saving control objective function:

[0092]

[0093] in:

[0094] θ represents the solution result of the robust energy-saving control objective function, that is, the energy-saving control strategy of the subway central air conditioning in the next moment;

[0095] This represents the cooling load provided by the chiller unit after adjusting the control parameters based on θ.

[0096] This represents the operating power of the cooling tower fan after adjusting the control parameters based on θ.

[0097] This represents the operating power of the water pump after adjusting the control parameters based on θ.

[0098] Constructing constraints for the robust energy-saving control objective function:

[0099]

[0100] in:

[0101] Indicates will The input is given to the subway central air conditioning temperature simulation model, and the model outputs the air conditioning supply temperature, M. min M represents the preset minimum air conditioning supply temperature. max This indicates the maximum preset air conditioning supply temperature.

[0102] Optionally, step S5 involves collecting short-term energy consumption time-series data of the subway central air conditioning system and optimizing the constructed robust energy-saving control objective function, including:

[0103] Collect current short-term energy consumption time series data for subway central air conditioning in Indicates t r Time to t R Short-term energy consumption time-series data at each moment is used to optimize the robust energy-saving control objective function based on the collected data. The optimization solution process is as follows:

[0104] S51: Will The data is input into the time series prediction model for energy consumption of the subway central air conditioning system to obtain t. R+1 Predicted energy consumption at any time

[0105] S52: Will Input the data into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation results.

[0106] S53: Set the reward value function V(θ):

[0107]

[0108] in:

[0109] V(θ) represents the reward value for adopting the energy-saving control strategy θ;

[0110] S54: Constructing the strategy control function Q(θ) based on the robust energy-saving control objective function, constraints, and reward value function V(θ):

[0111]

[0112] S55: The gradient descent algorithm is used to solve for θ in Q(θ) to obtain the solution result of the robust energy-saving control objective function. In this embodiment of the invention, the energy-saving control strategy θ represents the control result after energy-saving control of the control parameters of the energy-consuming equipment in the subway central air conditioning system, namely, the chilled water flow rate, chilled water supply temperature, chilled water return temperature provided by the chiller unit after control, the operating frequency of the cooling tower fan, the flow volume of the water pump, the efficiency of the water pump, and the efficiency of the frequency converter.

[0113] Optionally, in step S5, the solution result of the robust energy-saving control objective function is used as the energy-saving control strategy for the subway central air conditioning system in the next moment, including:

[0114] The solution of the robust energy-saving control objective function is used as the energy-saving control strategy of the subway central air conditioning in the next moment. Based on the energy-saving control strategy of the subway central air conditioning in the next moment, the energy-consuming equipment in the subway central air conditioning is controlled, and the air supply temperature of the subway central air conditioning is adaptively adjusted to achieve adaptive energy-saving control of the subway central air conditioning.

[0115] To address the above problems, the present invention provides an electronic device, the electronic device comprising:

[0116] Memory, storing at least one instruction;

[0117] Communication interfaces enable communication between electronic devices; and

[0118] The processor executes the instructions stored in the memory to implement the robust central air conditioning energy-saving control method described above.

[0119] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the aforementioned robust central air conditioning energy-saving control method.

[0120] Compared with existing technologies, this invention proposes a highly robust energy-saving control method for central air conditioning systems, which has the following advantages:

[0121] First, this proposal suggests a robust energy-saving control model for subway central air conditioning. This model includes a time-series prediction model for subway central air conditioning energy consumption, a temperature simulation model for subway central air conditioning, and an energy consumption control model. The energy consumption control model is used to adjust the parameters of the subway central air conditioning system, thereby controlling its energy consumption. The controlled energy consumption includes the cooling load provided by the chiller units, the operating power of the cooling tower fans, and the operating power of the water pumps. Based on this robust energy-saving control model, the adaptive temperature control process of the subway central air conditioning system is formally represented. The formally represented adaptive temperature control process... The process involves: collecting short-term energy consumption time-series data of the subway central air conditioning system and inputting the collected results into the subway central air conditioning energy consumption time-series prediction model to obtain the energy consumption at the next moment; inputting the energy consumption at the next moment into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation results; based on the predicted energy consumption at the next moment and the corresponding subway central air conditioning temperature simulation results, solving the control parameters of the energy-consuming equipment in the subway central air conditioning system based on the highly robust energy-saving control target, and controlling the energy-consuming equipment in the subway central air conditioning system based on the solved control parameters, thereby realizing adaptive parameter adjustment and adaptive temperature adjustment of the subway central air conditioning system based on short-term energy consumption. This solution is based on short-term energy consumption time-series data of the energy-consuming equipment of the subway central air conditioning system. It uses a subway central air conditioning energy consumption time-series prediction model to predict the energy consumption of different energy-consuming equipment at the next moment. The predicted energy consumption is converted into air conditioning supply air temperature, realizing the mapping between short-term energy consumption time-series data and air conditioning supply air temperature. Then, with the goal of minimizing energy consumption, the air conditioning supply air temperature range is used as a constraint, and the temperature difference between the regulated temperature and the predicted temperature is used as the reward value to construct a strategy control function. Solving the solution yields the energy-saving control strategy that satisfies the constraints and minimizes the temperature difference. The smaller the temperature difference, the closer the regulated temperature is to the predicted temperature. The energy consumption prediction result is similar to the time-series characteristics of the short-term energy consumption time-series data, and the prediction result is relatively smooth. This means that the temperature difference after regulation is small compared to the previous temperature, avoiding large changes in air conditioning supply air temperature.

[0122] Meanwhile, this scheme proposes a parameter optimization solution method, setting the current number of iterations for parameter optimization to m, the maximum number of iterations to Max, the initial value of m to 1, generating 2N parameter vector solutions, and dividing the generated parameter vector solutions into two groups, each group including N parameter vector solutions, wherein the generation formula of the parameter vector solutions is:

[0123]

[0124] Where: rand(0,1) represents a random number between 0 and 1, β max This represents the upper boundary of the pre-defined vector of parameters to be optimized.

[0125] β min This represents the lower boundary of the pre-defined parameter vector to be optimized; This represents the nth parameter vector solution in the i-th group, where i = 1, 2, n ∈ [1, N]; and the iteration result of the nth parameter vector solution in the i-th group after the m-th iteration is:

[0126] Solution of parameter vector The fitness value of the parameter vector solution is obtained by inputting it into the training objective function. Construct the training process judgment threshold Temp:

[0127]

[0128] If Temp < 0.25, it indicates that the current stage is the exploration stage, and the two sets of parameter vector solutions are iteratively updated; otherwise, proceed to step S16, where the iterative update formula is:

[0129]

[0130] in: Represents the solution of the random parameter vector of the i-th group; This represents the iteration result of the nth parameter vector solution in the i-th group after the m-th iteration; let m = m + 1, and return to the current step; if Temp ≥ 0.25, it indicates that the current stage is competitive, and the two groups of parameter vector solutions are iteratively updated, where the iterative update formula is:

[0131]

[0132]

[0133]

[0134]

[0135] in: This represents the parameter vector solution with the smallest applicability value in group 1 up to the current iteration number. Denote the parameter vector solution with the minimum fitness value in the second group as of the current iteration number; if m < Max, then m = m + 1, and repeat the current step; otherwise, calculate the fitness values of all current parameter vector solutions, select the parameter vector solution with the minimum fitness value as the optimized parameter in the subway central air-conditioning energy consumption time series prediction model, and construct the subway central air-conditioning energy consumption time series prediction model based on the optimized parameter. This solution uses an optimization and solution strategy of two groups of parameter vector solutions competing against each other to solve the parameters to be solved in the subway central air-conditioning energy consumption time series prediction model. In the process of generating the initial solution, a generation strategy combined with random weights is used to generate a more extensive and more random initial solution, expanding the initial space of the parameter vector solution. In the initial stage of algorithm iteration, an exploratory solution update scheme combined with random solutions is used to expand the search range of the parameter vector solution and quickly solve the model parameters. In the later stage of algorithm iteration, the two groups of parameter vector solutions compete against each other to update the parameter vector solution. If the antagonism between the two is relatively large, the parameter update range is expanded to avoid falling into the local optimal solution and ensure the effectiveness of the solved parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0136] Figure 1 FIG. is a schematic flowchart of a highly robust central air-conditioning energy-saving control method provided by an embodiment of the present invention;

[0137] Figure 2 FIG. is a schematic structural diagram of an electronic device for implementing a highly robust central air-conditioning energy-saving control method provided by an embodiment of the present invention.

[0138] The implementation, functional features, and advantages of the present invention will be further described in conjunction with the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0139] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0140] An embodiment of the present application provides a highly robust central air-conditioning energy-saving control method. The execution subject of the highly robust central air-conditioning energy-saving control method includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the highly robust central air-conditioning energy-saving control method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0141] Embodiment 1:

[0142] S1: Construct a subway central air-conditioning energy consumption time series prediction model. The constructed model takes short-term energy consumption time series data as input and the next moment's energy consumption as output.

[0143] The S1 step involves constructing a time-series prediction model for the energy consumption of the subway central air conditioning system, including:

[0144] A time-series prediction model for the energy consumption of subway central air conditioning is constructed. This model takes short-term energy consumption time-series data as input and outputs the energy consumption at the next time step. The short-term energy consumption time-series data is in the following format:

[0145]

[0146] in:

[0147] Indicates t h Energy consumption of the central air conditioning system in the subway at all times. P c (t h ) represents t h

[0148] The cooling load provided by the chiller unit in the central air conditioning system of the subway at all times, P e (t h ) represents t h The operating power (P) of the cooling tower fan in the central air conditioning system of the subway at any given time. s (t h ) represents t h The operating power of the water pumps in the central air conditioning system of the subway at all times;

[0149] Indicates time t1 to t H Short-term energy consumption time-series data at any given moment will As the input to the time-series prediction model for energy consumption of subway central air conditioning, the model prediction output is t. H+1 Energy consumption of subway central air conditioning

[0150] Based on short-term energy consumption time series data As input, the energy consumption prediction process based on the time-series prediction model for subway central air conditioning is as follows:

[0151] Set the maximum number of prediction iterations to d. num :

[0152]

[0153] in:

[0154] num represents the number of moments in the short-term energy consumption time series data, and th1 represents the preset iteration threshold.

[0155] d minThis represents the expected number of prediction iterations set in advance;

[0156] Set the current prediction iteration number to d, with an initial value of 1;

[0157] Repeat the following prediction iteration formula until the d-th term is obtained. num The result of the dth iteration, and the dth iteration result. num The result of the next iteration is used as the predicted t. H+1 Energy consumption of subway central air conditioning

[0158]

[0159] L 3,d-1 =σ[w3(L 1,d-1 +L 2,d-1 )+b3]

[0160]

[0161]

[0162] in:

[0163] σ(·) represents the activation function;

[0164] w1, w2, w3, w4 represent weight parameters, and b1, b2, b3, b4 represent bias parameters;

[0165] P d This represents the result of the d-th iteration. This is the iteration matrix.

[0166] Step S1 involves optimizing the parameters in the time-series prediction model for the energy consumption of the subway central air conditioning system, including:

[0167] The parameters to be optimized in the time-series prediction model for the energy consumption of the subway central air conditioning system include weight parameters w1, w2, w3, w4 and bias parameters b1, b2, b3, b4. These parameters are constructed as a parameter vector [w1, w2, w3, w4, b1, b2, b3, b4]. The parameter optimization process is as follows:

[0168] S11: Training dataset for constructing a time-series prediction model for the energy consumption of subway central air conditioning:

[0169] data={(Y k ,y k |k∈[1,K]}

[0170] in:

[0171] Y kThis represents the k-th training data set in the training dataset `data`. Each training data set consists of short-term energy consumption time-series data of varying lengths. k Y represents k Actual energy consumption data at the next moment;

[0172] S12: Constructing the training objective function F(β) for the time-series prediction model of energy consumption of subway central air conditioning:

[0173]

[0174] in:

[0175] β represents the parameter vector to be optimized in the time series prediction model of energy consumption of subway central air conditioning;

[0176] Indicate Y k The input is fed into the β-based time series prediction model for energy consumption of subway central air conditioning, and the model outputs the predicted value.

[0177] S13: Set the current number of iterations for parameter optimization to m, and the maximum number of iterations to Max. The initial value of m is 1. Generate 2N parameter vector solutions, and divide the generated parameter vector solutions into two groups, each group containing N parameter vector solutions. The formula for generating the parameter vector solutions is:

[0178]

[0179] in:

[0180] rand(0,1) represents a random number between 0 and 1, β max This represents the upper boundary of the pre-defined parameter vector to be optimized, β. min This represents the lower boundary of the pre-defined parameter vector to be optimized;

[0181] Let i represent the nth parameter vector solution in the i-th group, where i = 1, 2, n ∈ [1, N];

[0182] The iteration result of the nth parameter vector solution in the i-th group after the m-th iteration is: Solution of parameter vector The fitness value of the parameter vector solution is obtained by inputting it into the training objective function.

[0183] S14: Construct the training process judgment threshold Temp:

[0184]

[0185] S15: If Temp < 0.25, it indicates that the current is in the exploration stage, and iterative updates are performed on the two groups of parameter vector solutions. Otherwise, it turns to step S16. The iterative update formula is as follows:

[0186]

[0187] Where:

[0188] represents the random parameter vector solution of the i-th group;

[0189] represents the iterative result of the n-th parameter vector solution in the i-th group after the m-th iteration;

[0190] Let m = m + 1, and return to step S15;

[0191] S16: If Temp ≥ 0.25, it indicates that the current is in the competition stage, and iterative updates are performed on the two groups of parameter vector solutions. The iterative update formula is as follows:

[0192]

[0193]

[0194]

[0195]

[0196] Where:

[0197] represents the parameter vector solution with the minimum fitness value in the first group up to the current iteration number, represents the parameter vector solution with the minimum fitness value in the second group up to the current iteration number;

[0198] If m < Max, then m = m + 1, and return to step S16. Otherwise, calculate the fitness values of all current parameter vector solutions, select the parameter vector solution with the minimum fitness value as the optimized parameter in the subway central air-conditioning energy consumption time series prediction model, and build the subway central air-conditioning energy consumption time series prediction model based on the optimized parameter.

[0199] S2: Build a subway central air-conditioning temperature simulation model. The built model takes energy consumption as the input and the air supply temperature of the air conditioner as the output.

[0200] In the S2 step of building the subway central air-conditioning temperature simulation model, it includes:

[0201] A temperature simulation model for subway central air conditioning was constructed based on the SVR algorithm. The constructed model takes energy consumption as input and air conditioning supply temperature as output, that is, it establishes a hyperplane mapping relationship between energy consumption and air conditioning supply temperature.

[0202] M(P) = wP + b

[0203] in:

[0204] wP+b represents the constructed hyperplane, P represents energy consumption, M(P) represents the air conditioning supply temperature corresponding to P, w represents the linear weight of the hyperplane wP+b, and b represents the bias value of the hyperplane wP+b.

[0205] Training data was acquired, and a mean squared error loss function was established based on the training data. The particle swarm optimization algorithm was used to optimize and solve the hyperplane parameters w and b. Based on the solved hyperplane parameters, the hyperplane mapping relationship between energy consumption and air conditioning supply temperature in the subway central air conditioning temperature simulation model was constructed.

[0206] S3: Construct a robust energy-saving control model for subway central air conditioning. The constructed robust energy-saving control model for subway central air conditioning includes a time-series prediction model for subway central air conditioning energy consumption, a temperature simulation model for subway central air conditioning, and an air conditioning energy consumption control model. The adaptive temperature control process of the subway central air conditioning system is formally represented.

[0207] The S3 step involves constructing a robust energy-saving control model for the subway central air conditioning system, including:

[0208] A robust energy-saving control model for subway central air conditioning is constructed. This model includes a time-series prediction model for subway central air conditioning energy consumption, a temperature simulation model for subway central air conditioning, and an energy consumption control model. The energy consumption control model is used to adjust the parameters of the subway central air conditioning system, thereby controlling its energy consumption. The controlled energy consumption includes the cooling load provided by the chiller units, the operating power of the cooling tower fans, and the operating power of the water pumps. The corresponding energy consumption control formula is as follows:

[0209] P c =cq co (W1-W2)

[0210]

[0211]

[0212] in:

[0213] c represents the specific heat capacity of water;

[0214] q coThis indicates the chilled water flow rate supplied by the chiller unit, W1 represents the chilled water supply temperature, W2 represents the chilled water return temperature, and P... c q represents the cooling load provided by the chiller units in the central air conditioning system of the subway. co W1 and W2 represent the control parameters of the chiller unit;

[0215] f1 represents the operating frequency of the cooling tower fan, and f0 represents the rated frequency of the cooling tower fan. P represents the rated power of the cooling tower fan. e This indicates the operating power of the cooling tower fan in the subway central air conditioning system, and f1 represents the control parameters of the cooling tower fan.

[0216] ρ represents the density of water, g represents the acceleration due to gravity, and H represents the acceleration due to gravity. s This indicates the head of the water pump in the subway's central air conditioning system;

[0217] V represents the flow rate of the water pump, ρ s ρ represents the efficiency of a water pump. m P represents the efficiency of the frequency converter. s V,ρ represents the operating power of the water pump in the subway central air conditioning system. s ,ρ m These represent the control parameters of the water pump;

[0218] By adjusting the energy-consuming equipment parameters in the subway central air conditioning system in real time, the energy consumption after adjustment is obtained. The energy-consuming equipment includes chillers, cooling tower fans, and water pumps.

[0219] In step S3, the adaptive temperature control process of the subway central air conditioning system is formally represented based on the robust energy-saving control model of the subway central air conditioning system, including:

[0220] Based on the robust energy-saving control model of subway central air conditioning, the adaptive temperature control process of the subway central air conditioning system is formally represented. The formally represented adaptive temperature control process is as follows:

[0221] S31: Collect the current short-term energy consumption time series data of the subway central air conditioning system, and input the collected results into the subway central air conditioning energy consumption time series prediction model to obtain the energy consumption at the next moment.

[0222] S32: Input the energy consumption at the next moment into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation results;

[0223] S33: Based on the predicted energy consumption at the next moment and the corresponding simulation results of the subway central air conditioning temperature, the control parameters of the energy-consuming equipment in the subway central air conditioning are solved based on the high robust energy-saving control target, and the energy-consuming equipment in the subway central air conditioning is controlled based on the solved control parameters, so as to realize the adaptive parameter adjustment and adaptive temperature adjustment of the subway central air conditioning based on short-term energy consumption.

[0224] S4: Construct a robust energy-saving control objective function based on the robust energy-saving control model of subway central air conditioning.

[0225] The construction of the robust energy-saving control objective function in step S4 includes:

[0226] Construct a robust energy-saving control objective function:

[0227]

[0228] in:

[0229] θ represents the solution result of the robust energy-saving control objective function, that is, the energy-saving control strategy of the subway central air conditioning in the next moment;

[0230] This represents the cooling load provided by the chiller unit after adjusting the control parameters based on θ.

[0231] This represents the operating power of the cooling tower fan after adjusting the control parameters based on θ.

[0232] This represents the operating power of the water pump after adjusting the control parameters based on θ.

[0233] Constructing constraints for the robust energy-saving control objective function:

[0234]

[0235] in:

[0236] Indicates will The input is given to the subway central air conditioning temperature simulation model, and the model outputs the air conditioning supply temperature, M. min M represents the preset minimum air conditioning supply temperature. max This indicates the maximum preset air conditioning supply temperature.

[0237] S5: Collect the current short-term energy consumption time series data of the subway central air conditioning, optimize and solve the constructed robust energy-saving control objective function, and obtain the energy-saving control strategy of the subway central air conditioning in the next moment.

[0238] Step S5 involves collecting short-term energy consumption time-series data of the subway central air conditioning system and optimizing the constructed robust energy-saving control objective function, including:

[0239] Collect current short-term energy consumption time series data for subway central air conditioning in Indicates t r Time to t R Short-term energy consumption time-series data at each moment is used to optimize the robust energy-saving control objective function based on the collected data. The optimization solution process is as follows:

[0240] S51: Will The data is input into the time series prediction model for energy consumption of the subway central air conditioning system to obtain t. R+1 Predicted energy consumption at any time

[0241] S52: Will Input the data into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation results.

[0242] S53: Set the reward value function V(θ):

[0243]

[0244] in:

[0245] V(θ) represents the reward value for adopting the energy-saving control strategy θ;

[0246] S54: Constructing the strategy control function Q(θ) based on the robust energy-saving control objective function, constraints, and reward value function V(θ):

[0247]

[0248] S55: Solve for θ in Q(θ) using the gradient descent algorithm to obtain the solution of the robust energy-saving control objective function.

[0249] In step S5, the solution result of the robust energy-saving control objective function is used as the energy-saving control strategy for the subway central air conditioning system in the next moment, including:

[0250] The solution of the robust energy-saving control objective function is used as the energy-saving control strategy of the subway central air conditioning in the next moment. Based on the energy-saving control strategy of the subway central air conditioning in the next moment, the energy-consuming equipment in the subway central air conditioning is controlled, and the air supply temperature of the subway central air conditioning is adaptively adjusted to achieve adaptive energy-saving control of the subway central air conditioning.

[0251] Example 2:

[0252] likeFigure 2 The diagram shown is a structural schematic of an electronic device for implementing a highly robust central air conditioning energy-saving control method according to an embodiment of the present invention.

[0253] The electronic device 1 may include a processor 10, a memory 11, a communication interface 13 and a bus, and may also include a computer program, such as program 12, stored in the memory 11 and executable on the processor 10.

[0254] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 can include both internal and external storage units of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of program 12, but also to temporarily store data that has been output or will be output.

[0255] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (such as program 12 for implementing energy-saving control of central air conditioning) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0256] The communication interface 13 may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices, and to enable communication between internal components of the electronic device.

[0257] The bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0258] Figure 2 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 2 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0259] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0260] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0261] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0262] The program 12 stored in the memory 11 of the electronic device 1 is a combination of multiple instructions, which, when run in the processor 10, can achieve the following:

[0263] A robust energy-saving control model for subway central air conditioning is constructed, which includes a time-series prediction model for subway central air conditioning energy consumption, a temperature simulation model for subway central air conditioning, and an air conditioning energy consumption control model. The adaptive temperature control process of the subway central air conditioning system is formally represented.

[0264] A robust energy-saving control objective function is constructed based on the robust energy-saving control model of subway central air conditioning.

[0265] Collect short-term energy consumption time-series data of the subway central air conditioning system, optimize the constructed robust energy-saving control objective function, and obtain the energy-saving control strategy of the subway central air conditioning system in the next moment.

[0266] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 2 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0267] It should be noted that the sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, apparatus, article, or method. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0268] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0269] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A high-robust central air conditioning energy-saving control method, characterized in that, The method comprises: S1: constructing a metro central air conditioner energy consumption time sequence prediction model, the constructed model taking short-term energy consumption time sequence data as input and next moment energy consumption as output; The to-be-optimized parameters in the metro central air conditioning energy consumption time sequence prediction model include weight parameters and bias parameters The to-be-optimized parameters are constructed into a to-be-optimized parameter vector , wherein the parameter optimization process is: S11: constructing a training data set data of the metro central air conditioner energy consumption time sequence prediction model: Wherein: denotes the k-th group of training data in the training data set data, each group of training data being short-term energy consumption time series data of unequal length, denotes actual energy consumption data at the next time instant; S12: Constructing a training objective function of a metro central air conditioning energy consumption time sequence prediction model : Wherein: a vector of to-be-optimized parameters representing the subway central air conditioning energy consumption time sequence prediction model; representing the input into a subway central air conditioning energy time series prediction model based on predicted value of the model output S13: setting the iteration number of the current parameter optimization as m, the maximum iteration number as Max, the initial value of m as 1, generating 2N parameter vector solutions, and dividing the generated parameter vector solutions into two groups, each group including N parameter vector solutions, wherein the generation formula of the parameter vector solution is: Wherein: represents a random number between 0 and 1, represents a pre-set upper boundary of the parameter vector to be optimized, represents a pre-set lower boundary of the parameter vector to be optimized; represents the generated nth parameter vector solution in the ith set, , ; where the iteration result of the nth parameter vector solution in the ith group after the mth iteration is , the parameter vector solution is input into the training target function to obtain the fitness value of the parameter vector solution ; S14: Constructing a training process judgment threshold : S15: If , it means that the current is in the exploration phase, and the two sets of parameter vectors are updated iteratively, otherwise turn to step S16, where the iterative update formula is: Wherein: represents the random parameter vector solution of the i-th group; denotes the iteration result of the n-th parameter vector solution in the i-th group after the m-th iteration; Let , return to step S15; S16: If , indicates that it is currently in the competition phase, and iteratively updates the two sets of parameter vectors, where the iterative update formula is: Wherein: denotes the parameter vector solution with the minimum fitness value in the first group up to the current iteration number, denotes the parameter vector solution with the minimum fitness value in the second group up to the current iteration number; If , then , return to step S16, otherwise, calculate the fitness value of all current parameter vector solutions, select the parameter vector solution with the minimum fitness value as the optimization parameter in the metro central air conditioning energy consumption time series prediction model, and construct the metro central air conditioning energy consumption time series prediction model based on the optimization parameter; S2: constructing a metro central air conditioner temperature simulation model, the constructed model taking energy consumption as input and air conditioner supply air temperature as output; S3: constructing a metro central air conditioner robust energy-saving regulation and control model, the constructed metro central air conditioner robust energy-saving regulation and control model including the metro central air conditioner energy consumption time sequence prediction model, the metro central air conditioner temperature simulation model and the air conditioner energy consumption regulation and control model, and formally representing the adaptive temperature control regulation process of the metro central air conditioner system; S4: constructing a robust energy-saving regulation and control objective function according to the metro central air conditioner robust energy-saving regulation and control model; S5: collecting the current short-term energy consumption time sequence data of the metro central air conditioner, optimizing and solving the constructed robust energy-saving regulation and control objective function, and obtaining the energy-saving regulation and control strategy of the metro central air conditioner at the next moment.

2. The high-robust energy-saving control method for central air conditioning according to claim 1, wherein The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: representing the energy consumption of the subway central air conditioning at the moment, , representing the cooling load provided by the water chilling unit in the subway central air conditioning at the moment, representing the operating power of the cooling tower fan in the subway central air conditioning at the moment, representing the operating power of the water pump in the subway central air conditioning at the moment; express Time to Short-term energy consumption time-series data at any given moment will As input to the time-series prediction model for energy consumption of subway central air conditioning, the model prediction output is obtained. Energy consumption of subway central air conditioning ; Short-term energy consumption time series data The energy consumption prediction process based on the subway central air conditioning energy consumption time series prediction model takes the short-term energy consumption time series data as input. The maximum number of prediction iterations is set to : The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: represents the number of time points in the short-term energy consumption time series data, represents a pre-set iteration number threshold value, represents a pre-set prediction iteration number expectation value; The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The following prediction iteration formula is repeated until the result of the first iteration is obtained, and the result of the first iteration is taken as the predicted energy consumption of the subway central air conditioning at the moment The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: denotes an activation function; denotes a weight parameter, denotes a bias parameter; represents the result of the dth iteration, , is the iteration matrix.

3. The high-robust energy-saving control method for central air conditioning according to claim 1, wherein The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: represents a constructed hyperplane, P represents energy consumption, represents an air conditioning supply air temperature corresponding to P, w represents a linear weight of the hyperplane , and b represents a bias value of the hyperplane ; The training data is acquired, and a mean square error loss function is established based on the training data Optimization is performed, and a hyperplane mapping relationship between energy consumption and air conditioning supply air temperature in the subway central air conditioning temperature simulation model is constructed based on the obtained hyperplane parameters.

4. The high-robust energy-saving control method for central air conditioning according to claim 1, wherein The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: Qch represents the chilled water flow rate provided by the chiller unit, Tch_in represents the chilled water supply temperature, Tch_out represents the chilled water return temperature, Qch represents the cooling load provided by the chiller unit in the subway central air conditioning system, Qch represents the control parameter of the chiller unit; represents the operating frequency of the cooling tower fan, represents the rated frequency of the cooling tower fan, represents the rated power of the cooling tower fan, represents the operating power of the cooling tower fan in the subway central air conditioning, represents the control parameter of the cooling tower fan; denotes the density of water, g denotes the acceleration of gravity, denotes the head of the water pump in the subway central air conditioning; represents the flow volume of the water pump, represents the efficiency of the water pump, represents the efficiency of the frequency converter, represents the operating power of the water pump in the subway central air conditioning, represents the control parameter of the water pump; The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air conditioner energy consumption time sequence prediction model comprises: The step S1 of constructing the metro central air 5. The high-robust energy-saving control method for central air conditioning according to claim 4, characterized in that, The S3 step formally represents the adaptive temperature control adjustment process of the subway central air conditioning system based on the subway central air conditioning robust energy-saving regulation and control model, including: The adaptive temperature control adjustment process is formally represented based on the subway central air conditioning robust energy-saving regulation and control model, wherein the adaptive temperature control adjustment process after formal representation is: S31: Collect the current short-term energy consumption time series data of the subway central air conditioning, and input the collection result into the subway central air conditioning energy consumption time series prediction model to obtain the energy consumption at the next time; S32: Input the energy consumption at the next time into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation result; S33: According to the predicted energy consumption at the next time and the corresponding subway central air conditioning temperature simulation result, the regulation and control parameters of the energy-consuming equipment in the subway central air conditioning are solved based on the high-robust energy-saving regulation and control target, and the energy-consuming equipment in the subway central air conditioning is regulated and controlled based on the solved regulation and control parameters.

6. The high-robust energy-saving control method for central air conditioning according to claim 5, wherein The S4 step includes constructing a robust energy-saving regulation and control objective function, including: The robust energy-saving regulation and control objective function is constructed: Wherein: represents the solution result of the robust energy-saving regulation objective function, i.e. the energy-saving regulation strategy of the subway central air conditioner at the next time point; indicates based on the cold load provided by the water chiller after the regulation of the regulation parameter indicates based on The running power of the cooling tower fan after the regulation and control of the regulation and control parameters in the cooling tower fan; representing based on the operating power of the water pump after the regulating parameter in the water pump is regulated; The constraint condition for constructing the robust energy-saving regulation and control objective function is: Wherein: represents that input to a subway central air conditioning temperature simulation model, an air conditioning supply air temperature output from the model, represents a pre-set air conditioning supply air temperature minimum value, represents a pre-set air conditioning supply air temperature maximum value.

7. The high-robust energy-saving control method for central air conditioning according to claim 1, wherein, The S5 step includes collecting the current short-term energy consumption time series data of the subway central air conditioning, and optimizing and solving the constructed robust energy-saving regulation and control objective function, including: Collect current short-term energy consumption time series data for subway central air conditioning ,in express Time to Short-term energy consumption time-series data at each moment is used to optimize the robust energy-saving control objective function based on the collected data. The optimization solution process is as follows: S51: inputting the data to the subway central air conditioning energy consumption time sequence prediction model to obtain the predicted value of the energy consumption at the time point ;​ S52: obtaining The input is input into the subway central air conditioning temperature simulation model to obtain the corresponding subway central air conditioning temperature simulation result ; S53: Set the reward value function : Wherein: represents a reward value for employing an energy saving control strategy ​ S54: constructing a strategy regulation function based on the robust energy-saving regulation target function, the constraint condition and the reward value function constructing a strategy regulation function : S55: Using the gradient descent algorithm to... In The solution is obtained by solving the problem and obtaining the solution result of the robust energy-saving control objective function.

8. The high-robust energy-saving control method for central air conditioning according to claim 7, characterized in that, The S5 step includes taking the solution of the robust energy-saving regulation and control objective function as the energy-saving regulation and control strategy of the subway central air conditioning at the next time, including: The solution of the robust energy-saving regulation and control objective function is taken as the energy-saving regulation and control strategy of the subway central air conditioning at the next time, and the energy-consuming equipment in the subway central air conditioning is regulated and controlled based on the energy-saving regulation and control strategy of the subway central air conditioning at the next time, thereby adaptively adjusting the supply air temperature of the subway central air conditioning.

Images