阅读说明：本技术 一种冰蓄冷系统的节能运行方法、装置、设备及存储介质 (Energy-saving operation method, device and equipment of ice storage system and storage medium ) 是由 吴健申 倪伟 伍兰昌 张岳 范健桦 于 2019-10-11 设计创作，主要内容包括：本发明公开了一种冰蓄冷系统的节能运行方法、装置、设备及存储介质,包括获取n个历史样本数据,基于PCA算法对所述历史样本数据进行降维,得到对应k个特征向量的降维数据集,基于随机梯度下降回归算法处理所述降维数据集,获得所述输入向量中各分量的最优权重,根据所述输出参数和不同时间段的电价向量,得到电价计算模型,从而得到冷水机组制冰和制冷的运行时间段。本发明实施例基于PCA算法对多维的历史样本数据进行降维处理,降低随机梯度下降回归算法的输入数据的维度,减少系统计算量,根据随机梯度下降回归算法寻求最佳权重从而得到冷水机组制冰和制冷分别运行的时间段,减少运营成本和能源浪费。(The invention discloses an energy-saving operation method, device, equipment and storage medium of an ice storage system, which comprises the steps of obtaining n pieces of historical sample data, carrying out dimensionality reduction on the historical sample data based on a PCA algorithm to obtain a dimensionality reduction data set corresponding to k feature vectors, processing the dimensionality reduction data set based on a stochastic gradient descent regression algorithm to obtain the optimal weight of each component in an input vector, and obtaining an electricity price calculation model according to output parameters and electricity price vectors in different time periods so as to obtain the ice making and refrigerating operation time periods of a water chilling unit. The embodiment of the invention performs dimensionality reduction processing on multi-dimensional historical sample data based on the PCA algorithm, reduces the dimensionality of input data of the random gradient descent regression algorithm, reduces the system computation amount, seeks the optimal weight according to the random gradient descent regression algorithm so as to obtain time periods for ice making and refrigeration of the water chilling unit to respectively operate, and reduces the operation cost and energy waste.)

1. An energy-saving operation method of an ice storage system is characterized by comprising the following steps:

acquiring n historical sample data, wherein each historical sample data comprises an input vector representing a related parameter of the refrigerating capacity and an output parameter representing the refrigerating capacity on the next day, and n is a positive integer;

performing dimensionality reduction on the historical sample data based on a PCA algorithm to obtain a dimensionality reduction dataset corresponding to k eigenvectors, wherein k is a positive integer smaller than n;

processing the dimensionality reduction data set based on a stochastic gradient descent regression algorithm to obtain the optimal weight of each component in the input vector, wherein the output parameter is represented as the weighted average of the input vector;

obtaining a power price calculation model according to the output parameters and the power price vectors in different time periods;

and obtaining the ice making and refrigerating operation time period of the water chilling unit according to the input vector of the current day and the electricity price calculation model.

2. The energy-saving operation method of the ice thermal storage system according to claim 1, characterized in that:

the reducing dimension of the historical sample data based on the PCA algorithm to obtain a dimension-reduced data set corresponding to the k eigenvectors comprises the following steps:

calculating the mean value of the historical sample data, and performing data centralization on the historical sample data based on the mean value;

calculating a covariance matrix of the historical sample data after data centralization;

calculating eigenvalues and corresponding eigenvectors of the covariance matrix;

arranging the eigenvalues from large to small, selecting the largest k of the eigenvalues, and arranging the eigenvectors corresponding to the eigenvalues of the selected k into an eigenvector matrix, wherein k is a positive integer smaller than n;

and projecting the mean value onto the feature vectors to form a dimensionality reduction data set corresponding to the k feature vectors.

3. The energy-saving operation method of the ice thermal storage system according to claim 2, characterized in that:

the sum of the k eigenvalues is chosen to be greater than or equal to 90% of the sum of the n eigenvalues.

4. The energy-saving operation method of the ice thermal storage system according to claim 1, characterized in that:

the processing the dimensionality reduction data set based on the stochastic gradient descent regression algorithm to obtain the optimal weight of each component in the input vector comprises the following steps:

preprocessing the dimensionality reduction data set and eliminating data abnormal values;

normalizing the values in the preprocessed dimension reduction dataset;

the input vector is x_i＝(x₁，x₂，x₃，...，x_m) Wherein m is a positive integer, m represents that the parameters related to the refrigerating capacity are m, and the parameters correspond to x based on the linear relation between the refrigerating capacity and the parameters related to the refrigerating capacity_iPredicted cooling capacity h_θ(x) Watch (A)Shown as follows:

wherein θ ═ θ₀，θ₁，θ₂，...，θ_n) Is a weight vector expressed as the components of the input vector to the predicted cooling capacity h_θ(x) The degree of influence of (c);

the loss function is defined as:

and y represents an output parameter corresponding to the input vector x in the historical sample data, theta is assigned, the partial derivative of the loss function is calculated, J (theta) approaches to the minimum value according to an updating formula of theta, and the optimal theta value is calculated.

5. The energy-saving operation method of the ice thermal storage system according to claim 1, characterized in that: the obtaining of the electricity price calculation model according to the output parameters and the electricity price vectors in different time periods comprises:

output parameter based on optimal weight, i.e. predicted cooling capacity h_θ(x) Is shown as

h_θ(x)＝Q＝F*(Q1*t1+Q2*t2)，

Wherein F is the number of the running water chilling units, Q1 is the refrigerating capacity of each water chilling unit per hour, t1 is the refrigerating running time, Q2 is the equivalent refrigerating capacity of the water chilling units per hour for ice making, and t2 is the ice making running time;

the electricity price vectors in different time periods are represented as P ═ P1, P2, P3,.. so, pl ], l is a positive integer, the refrigeration vectors in different time periods each day are represented as c1 ═ Q1, Q2, Q3,. so, ql ], the ice making vectors in different time periods each day are represented as c2 ═ d1, d2, d3,. so, dl ], then Q1 ═ Q1+ Q2+ Q3+.. + ql, Q2 ═ d1+ d2+ d3+. so + dl, then the electricity price calculation model is: p (Q1, Q2) ═ P (c1+ c 2).

6. An energy-saving operation device of an ice storage system is characterized by comprising:

the system comprises an acquisition unit, a storage unit and a control unit, wherein the acquisition unit is used for acquiring n historical sample data, each historical sample data comprises an input vector representing a related parameter of the refrigerating capacity and an output parameter representing the refrigerating capacity on the next day, and n is a positive integer;

a PCA algorithm unit, configured to perform dimensionality reduction on the historical sample data based on a PCA algorithm to obtain a dimensionality reduction dataset corresponding to k eigenvectors, where k is a positive integer smaller than n;

a stochastic gradient descent regression algorithm unit, configured to process the dimensionality reduction data set based on a stochastic gradient descent regression algorithm to obtain an optimal weight of each component in the input vector, where the output parameter is represented as a weighted average of the input vector;

the electricity price calculation model unit is used for obtaining an electricity price calculation model according to the output parameters and the electricity price vectors in different time periods;

and the output unit is used for obtaining the ice making and refrigerating operation time period of the water chilling unit according to the input vector of the current day and the electricity price calculation model.

7. An energy-saving operation equipment of ice cold storage system, its characterized in that: comprises at least one control processor and a memory for communicative connection with the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform the energy efficient method of operating an ice storage system as claimed in any one of claims 1 to 5.

8. A computer-readable storage medium characterized by: the computer-readable storage medium stores computer-executable instructions for causing a computer to perform the method for energy efficient operation of an ice thermal storage system as claimed in any one of claims 1 to 5.

Technical Field

The invention relates to the field of operation control of ice cold storage central air-conditioning systems, in particular to an energy-saving operation method, device, equipment and storage medium of an ice cold storage system.

Background

In the actual operation process of the ice storage central air conditioning system at the present stage, water is made into ice by utilizing the time interval of low electricity price at the off-peak time at night and stored in an ice storage tank, and the ice is dissolved and refrigerated to supply cold to a terminal at the peak electricity utilization time in the daytime, so that the refrigeration cost is greatly reduced, but the energy loss and the production cost are still very high when the actual system is operated, mainly because the operation quantity of a water chilling unit and the ice making quantity of the unit are determined by operators according to experience in the ice storage time interval, the refrigeration demand of the next day is predicted only by manpower, data is absent, randomness is strong, insufficient ice making quantity or over saturation is easily caused when the central air conditioning is used for the second day, and energy waste is caused.

Disclosure of Invention

The invention aims to at least solve one of the technical problems in the prior art, and provides an energy-saving operation method, device, equipment and storage medium of an ice storage system.

A first aspect of an embodiment of the present invention provides an energy-saving operation method of an ice thermal storage system, including

Acquiring n historical sample data, wherein each historical sample data comprises an input vector representing a related parameter of the refrigerating capacity and an output parameter representing the refrigerating capacity on the next day, and n is a positive integer;

performing dimensionality reduction on the historical sample data based on a PCA algorithm to obtain a dimensionality reduction dataset corresponding to k eigenvectors, wherein k is a positive integer smaller than n;

processing the dimensionality reduction data set based on a stochastic gradient descent regression algorithm to obtain the optimal weight of each component in the input vector, wherein the output parameter is represented as the weighted average of the input vector;

obtaining a power price calculation model according to the output parameters and the power price vectors in different time periods;

and obtaining the ice making and refrigerating operation time period of the water chilling unit according to the input vector of the current day and the electricity price calculation model.

The energy-saving operation method based on the ice storage system at least has the following beneficial effects: the method comprises the steps of carrying out dimensionality reduction processing on multi-dimensional historical sample data based on a PCA algorithm, reducing dimensionality of input data of a random gradient descent regression algorithm, reducing system calculation amount, seeking optimal weight according to the random gradient descent regression algorithm to obtain time periods when a water chilling unit respectively carries out ice making and refrigeration, and accordingly obtaining an electricity price calculation model based on the optimal weight, and operators can obtain an operation mode and a total electricity price of the water chilling unit on the next day by inputting refrigeration capacity related parameters.

According to the energy-saving operation method of the ice storage system in the first aspect of the embodiment of the present invention, the dimensionality reduction is performed on the historical sample data based on the PCA algorithm to obtain a dimensionality reduction dataset corresponding to k eigenvectors, including:

calculating the mean value of the historical sample data, and performing data centralization on the historical sample data based on the mean value;

calculating a covariance matrix of the historical sample data after data centralization;

calculating eigenvalues and corresponding eigenvectors of the covariance matrix;

arranging the eigenvalues from large to small, selecting the largest k of the eigenvalues, and arranging the eigenvectors corresponding to the eigenvalues of the selected k into an eigenvector matrix, wherein k is a positive integer smaller than n;

and projecting the mean value onto the feature vectors to form a dimensionality reduction data set corresponding to the k feature vectors.

According to the energy-saving operation method of the ice storage system in the first aspect of the embodiment of the invention, the sum of the k characteristic values is greater than or equal to 90% of the sum of the n characteristic values.

According to the energy-saving operation method of the ice storage system in the first aspect of the embodiment of the present invention, the processing the dimensionality reduction data set based on the stochastic gradient descent regression algorithm to obtain the optimal weight of each component in the input vector includes:

preprocessing the dimensionality reduction data set and eliminating data abnormal values;

normalizing the values in the preprocessed dimension reduction dataset;

the input vector is x_i＝(x₁,x₂,x₃,…,x_m) And i is 0,1,2,3, …, n, wherein m is a positive integer and represents that the parameters related to the refrigerating capacity are m, and the parameters related to the refrigerating capacity are in a linear relation based on the refrigerating capacity and correspond to x_iPredicted cooling capacity h_θ(x) Is shown as

Wherein θ ═ θ₀,θ₁,θ₂,…,θ_n) Is a weight vector expressed as the components of the input vector to the predicted cooling capacity h_θ(x) The degree of influence of (c);

the loss function is defined as:

and y represents an output parameter corresponding to the input vector x in the historical sample data, theta is assigned, the partial derivative of the loss function is calculated, J (theta) approaches to the minimum value according to an updating formula of theta, and the optimal theta value is calculated.

According to the energy-saving operation method of the ice storage system in the first aspect of the embodiment of the present invention, the obtaining of the electricity price calculation model according to the output parameters and the electricity price vectors in different time periods includes:

output parameter based on optimal weight, i.e. predicted cooling capacity h_θ(x) Is shown as

h₆(x)＝Q＝F*(Q1*t1+Q2*t2)，

Wherein F is the number of the running water chilling units, Q1 is the refrigerating capacity of each water chilling unit per hour, t1 is the refrigerating running time, Q2 is the equivalent refrigerating capacity of the water chilling units per hour for ice making, and t2 is the ice making running time;

the electricity price vectors in different time periods are represented as P ═ P1, P2, P3, …, pl ], l is a positive integer, the refrigeration vectors in different time periods per day are represented as c1 ═ Q1, Q2, Q3, …, ql, the ice making vectors in different time periods per day are represented as c2 ═ d1, d2, d3, …, dl ], then Q1 ═ Q1+ Q2+ Q3+ … + ql, Q2 ═ d1+ d2+ d3+ … + dl, and then the electricity price calculation model is:

P(Q1,Q2)＝P*(c1+c2)。

a second aspect of an embodiment of the present invention provides an energy-saving operation device of an ice thermal storage system, including:

the system comprises an acquisition unit, a storage unit and a control unit, wherein the acquisition unit is used for acquiring n historical sample data, each historical sample data comprises an input vector representing a related parameter of the refrigerating capacity and an output parameter representing the refrigerating capacity on the next day, and n is a positive integer;

a PCA algorithm unit, configured to perform dimensionality reduction on the historical sample data based on a PCA algorithm to obtain a dimensionality reduction dataset corresponding to k eigenvectors, where k is a positive integer smaller than n;

a stochastic gradient descent regression algorithm unit, configured to process the dimensionality reduction data set based on a stochastic gradient descent regression algorithm to obtain an optimal weight of each component in the input vector, where the output parameter is represented as a weighted average of the input vector;

the electricity price calculation model unit is used for obtaining an electricity price calculation model according to the output parameters and the electricity price vectors in different time periods;

and the output unit is used for obtaining the ice making and refrigerating operation time period of the water chilling unit according to the input vector of the current day and the electricity price calculation model.

A third aspect of embodiments of the present invention provides an energy-efficient operation apparatus of an ice thermal storage system, comprising at least one control processor and a memory for communicative connection with the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform the energy efficient method of operating an ice storage system as described in any one of the above.

A third aspect of embodiments of the present invention provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the method for energy efficient operation of an ice thermal storage system as described in any one of the above.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The invention is further described below with reference to the accompanying drawings and examples;

FIG. 1 is an overall flow chart of an embodiment of the present invention;

FIG. 2 is a flowchart illustrating the step S200 according to an embodiment of the present invention;

FIG. 3 is a flowchart illustrating the step S300 according to an embodiment of the present invention;

FIG. 4 is a flowchart illustrating the step S400 according to an embodiment of the present invention;

FIG. 5 is a block diagram of an apparatus according to an embodiment of the present invention;

fig. 6 is a connection diagram of the apparatus according to the embodiment of the present invention.

Detailed Description

Reference will now be made in detail to the present preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to like elements throughout.

In the description of the present invention, it should be understood that the orientation descriptions referred to, for example, the orientation or positional relationship indicated above and below, are based on the orientation or positional relationship shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the device or element referred to must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present invention.