CN117826900A

CN117826900A - Control method and device for cold and heat source system and computer readable storage medium

Info

Publication number: CN117826900A
Application number: CN202410024146.3A
Authority: CN
Inventors: 郝吉芳; 杨康
Original assignee: BOE Technology Group Co Ltd; Beijing BOE Technology Development Co Ltd
Current assignee: BOE Technology Group Co Ltd; Beijing BOE Technology Development Co Ltd
Priority date: 2024-01-05
Filing date: 2024-01-05
Publication date: 2024-04-05

Abstract

The present disclosure discloses a control method and apparatus for a cold and heat source system, a computer readable storage medium, and a computer device, including: performing parameter training by using a trained energy consumption relation model based on system parameters to obtain predicted electric quantity; taking the predicted electric quantity as a reference value, and acquiring a first class value and a second class value in the actual electric quantity of the cold and heat source system in a first preset time period; acquiring the distribution of the adjustable parameters corresponding to the first type of values to serve as first distribution, and acquiring the distribution of the adjustable parameters corresponding to the second type of values to serve as second distribution; and calculating a difference value between the first distribution and the second distribution, and when the difference value is larger than a preset threshold value, taking the corresponding adjustable parameter as an optimizable parameter, and determining an optimization control value of the optimizable parameter based on the first distribution of the optimizable parameter. In the embodiment, the adjustable parameters with large distribution difference values are optimized according to the distribution rule, so that the energy consumption of the system is reduced.

Description

Control method and device for cold and heat source system and computer readable storage medium

Technical Field

The disclosure relates to the field of debugging technologies, and in particular, to a control method and device of a cold and heat source system, a computer readable storage medium and computer equipment.

Background

At present, a plant cold and heat source system comprises a plurality of devices, the devices need to complete heat exchange or output cold/heat energy under the control of various parameters, each process needs to consume electric energy, and one cold and heat source system comprises up to hundreds of parameters, so that how to optimally control the system under the condition of meeting the operation requirement of the system and ensuring the stability of the system so as to enable the devices to work under the working condition of minimum energy consumption is a current urgent problem.

Disclosure of Invention

In order to solve at least one of the above problems, a first aspect of the present disclosure provides a control method of a cold and heat source system, comprising:

acquiring system parameters for electric quantity prediction;

acquiring predicted electric quantity by using a trained energy consumption relation model based on system parameters for electric quantity prediction;

taking the predicted electric quantity as a reference value, acquiring a first class value and a second class value in the actual electric quantity of the cold and heat source system in a first preset time period, wherein the value of the first class value is lower than the reference value, and the value of the second class value is higher than the reference value;

acquiring the distribution of the adjustable parameters corresponding to the first type of values to serve as first distribution, and acquiring the distribution of the adjustable parameters corresponding to the second type of values to serve as second distribution;

Calculating a difference value between the first distribution and the second distribution, and taking the corresponding adjustable parameter as an optimizable parameter when the difference value is larger than a preset threshold value;

an optimal control value for the optimizable parameter is determined based on the first distribution of the optimizable parameter.

Optionally, determining the optimal control value of the optimizable parameter based on the first distribution of the optimizable parameter further comprises:

calculating a distribution interval of the optimizable parameter within a preset probability range based on the first distribution of the optimizable parameter;

the distribution interval is used as a value interval of the optimized control value to determine the optimized control value of the optimized parameter.

Optionally, before obtaining the predicted electric quantity by using the trained energy consumption relation model based on the system parameters for electric quantity prediction, the method further comprises:

building an energy consumption relation model, wherein the built energy consumption relation model comprises energy consumption parameters;

acquiring a sample set based on system parameters for constructing an energy consumption relation model in a second preset time period, wherein the second preset time period is the same as or different from the first time period;

dividing a training set and a testing set based on the sample set;

training the energy consumption relation model based on a training set, and fitting energy consumption parameters by using a fitting curve formed by the training set to obtain the energy consumption relation model to be tested;

And testing the energy consumption relation model by using the test set to verify the trained energy consumption relation model.

Optionally, the constructed energy consumption relation model is: electric quantity=a1 ambient temperature P total cooling capacity + a2 total cooling capacity + C,

wherein the electric quantity represents the total electric consumption of the cold and heat source system, the ambient temperature represents the temperature of the space where the cold and heat source system is positioned, A1, A2, P and C are energy consumption parameters,

total coldness = Σ (return water temperature of each device-supply water temperature of each device) × water flow rate × fixed parameter.

Optionally, calculating the difference value between the first distribution and the second distribution further comprises:

based on the first distribution and the second distribution, JS divergence values are calculated, wherein the JS divergence values are difference values, and the calculation of the JS divergence values meets the following conditions:

wherein P (x) represents a first distribution, Q (x) represents a second distribution, and operator KL (A||B) represents

Optionally, the adjustable parameters include: one or more of a chilled water inlet temperature of the chiller, a chilled water outlet temperature of the chiller, a primary pump frequency, a secondary pump frequency, a cooling tower frequency, and an operational status.

Optionally, obtaining the sample set based on the system parameters for constructing the energy consumption relation model within the second preset time period further includes:

Obtaining system parameters for constructing an energy consumption relation model in a second preset time period,

dividing the second preset time period into a plurality of time windows;

sampling system parameters for constructing an energy consumption relation model in a second preset time period by taking a time window as a time unit to construct a sample set.

Optionally, the method further includes, after obtaining the distribution of the adjustable parameter corresponding to the first class value as a first distribution and obtaining the distribution of the adjustable parameter corresponding to the second class value as a second distribution:

a first distribution curve is drawn based on the first distribution and a second distribution curve is drawn based on the second distribution,

and drawing and storing the first distribution curve and the second distribution curve of the same adjustable parameter as the distribution curve of the adjustable parameter in the same coordinate system.

A second aspect of the present disclosure provides a control apparatus of a cold and heat source system, including:

the acquisition unit is used for acquiring system parameters for the energy consumption relation model;

the prediction unit is used for obtaining predicted electric quantity by utilizing the trained energy consumption relation model based on system parameters for the energy consumption relation model;

the classification unit is used for acquiring a first class value and a second class value in the actual electric quantity of the cold and heat source system in a first preset time period by taking the predicted electric quantity as a reference value, wherein the value of the first class value is lower than the reference value, and the value of the second class value is higher than the reference value;

The analysis unit is used for acquiring the distribution of the adjustable parameters corresponding to the first type of values to be used as first distribution, and acquiring the distribution of the adjustable parameters corresponding to the second type of values to be used as second distribution;

a calculation unit for calculating a difference value between the first distribution and the second distribution, and when the difference value is larger than a preset threshold value, taking the corresponding adjustable parameter as an optimizable parameter,

and an optimizing unit for determining an optimized control value of the optimizable parameter based on the first distribution of the optimizable parameter.

A third aspect of the present disclosure provides a computer-readable storage medium, having a computer program stored thereon,

the program, when executed by the processor, implements the control method of the cold heat source system as described above.

A fourth aspect of the present disclosure provides a computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor,

the processor implements the control method of the cold heat source system as described above when executing the program.

The beneficial effects of the present disclosure are as follows:

aiming at the existing problems at present, the control method and device, the computer readable storage medium and the computer equipment of the cold and heat source system are formulated, the energy consumption higher than the predicted electric quantity and the energy consumption higher than the predicted electric quantity are divided by taking the predicted electric quantity as a reference, the adjustable parameters with larger difference in distribution are determined for the two types of energy consumption, and the optimal control value is determined based on the probability distribution, so that the adjustable parameters with larger influence on the energy consumption can be determined, accurate parameter optimization is performed, the energy consumption of the system is reduced, the energy saving of the system is realized, and the system has wide application prospect.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings may be obtained according to these drawings without inventive effort for a person of ordinary skill in the art.

FIG. 1 illustrates an exemplary flowchart of a control method of a cold heat source system according to an embodiment of the present disclosure;

fig. 2 illustrates an architecture diagram of a cold heat source system to which a control method of the cold heat source system according to an embodiment of the present disclosure may be applied;

fig. 3 illustrates an exemplary flowchart of a control method of a cold heat source system according to an embodiment of the present disclosure;

FIG. 4 illustrates a visual plot of predicted values versus true values for an energy consumption relationship model constructed in accordance with an embodiment of the present disclosure;

FIG. 5 illustrates a distribution graph plotted against an exemplary tunable parameter of a first class of values and a second class of values;

FIG. 6 shows an effect graph of a dimension reduction visualization for all of the system parameters;

FIG. 7 shows an effect graph of dimension reduction visualization for parameters that differ among system parameters;

FIG. 8 shows a schematic block diagram of an optimization control device of a cold heat source system according to an embodiment of the present disclosure;

fig. 9 illustrates a schematic configuration of a computer device implementing a control method of the cold heat source system of the present disclosure.

Detailed Description

In order to more clearly illustrate the present disclosure, the present disclosure is further described below in connection with the preferred embodiments and the accompanying drawings. Like parts in the drawings are denoted by the same reference numerals. It is to be understood by persons of ordinary skill in the art that the following detailed description is illustrative and not restrictive, and should not be taken as limiting the scope of the present disclosure.

It should be noted that, in this disclosure, the terms "having," "including," "comprising," and the like are all open-ended, i.e., when a module is described as "having," "including," or "comprising" a first element, a second element, and/or a third element, it is intended that the module include other elements in addition to the first element, the second element, and/or the third element. In addition, ordinal numbers such as "first", "second", and "third" in this disclosure are not intended to limit a specific order, but merely to distinguish between the individual portions.

In order to solve one of the above problems, as shown in fig. 1, an embodiment of the present disclosure provides a control method of a heat and cold source system, including:

S1, acquiring system parameters for electric quantity prediction;

s2, acquiring predicted electric quantity by using a trained energy consumption relation model based on the system parameters for electric quantity prediction;

s3, taking the predicted electric quantity as a reference value, and acquiring a first class value and a second class value in the actual electric quantity of the cold and heat source system in a first preset time period, wherein the value of the first class value is lower than the reference value, and the value of the second class value is higher than the reference value;

s4, acquiring the distribution of the adjustable parameters corresponding to the first type of values to serve as first distribution, and acquiring the distribution of the adjustable parameters corresponding to the second type of values to serve as second distribution;

step S5, calculating a difference value between the first distribution and the second distribution, when the difference value is larger than a preset threshold value, taking the corresponding adjustable parameter as an optimizable parameter,

and S6, determining an optimal control value of the optimizable parameter based on the first distribution of the optimizable parameter.

In the embodiment, the energy consumption higher than the predicted electric quantity and the energy consumption higher than the predicted electric quantity are divided by taking the predicted electric quantity as a reference, the adjustable parameters with larger difference in distribution are determined for the two types of energy consumption, and the optimal control value is determined based on the probability distribution, so that the adjustable parameters with larger influence on the actual electric quantity can be determined, accurate parameter optimization can be performed, the energy consumption of a system is reduced, and the energy saving of the system is realized.

In order to facilitate understanding of the technical solution of the present embodiment, the following describes an actual architecture of the cold and heat source system applied by the method provided in the present embodiment with reference to fig. 2.

Referring to FIG. 2, a block diagram of an exemplary heat and cold source system is shown. As can be seen, the plant heat and cold source system includes a plurality of devices including a chiller, chilled water pumps or pipes, and a cooling tower. The water chilling unit comprises an evaporator, a compressor and a condenser to jointly complete the refrigeration process; the chilled water pump or pipeline provides chilled water for the evaporator, and the chilled water pump or pipeline completes water inlet and water outlet circulation for the water chilling unit; the cooling tower actually forms a cooling tower system with a fan which supplies outdoor ambient air to the cooling tower and outputs cooled air from the cooling tower for heat exchange, and a cooling water pump pumps cooling water to supply cooling water of the chiller to the cooling tower for cooling and outputs the cooling tower for circulation cooling. Although not shown, the chiller that supplies the cooling water includes a middle temperature chiller for supplying the middle temperature cooling water or the middle temperature chilled water and a low temperature chiller for supplying the low temperature cooling water or the low temperature chilled water.

Because the cold and heat source system is a system which is matched with a plurality of devices in a plurality of units to complete cold and heat exchange or heat energy generation, the system comprises up to hundreds of parameters, the system needs to control various parameters to complete the functions, and the energy consumption is the result of the combined action of the various parameters. For example, the system parameters include an electric quantity of each electric meter, an external environment temperature, a water flow rate, a water inlet temperature of chilled water of each chiller (including a water inlet temperature of medium-temperature chilled water and a water inlet temperature of low-temperature chilled water), a water outlet temperature of chilled water of each chiller (including a water outlet temperature of medium-temperature chilled water and a water outlet temperature of low-temperature chilled water), a water inlet temperature of cooling water of each chiller (including a water inlet temperature of medium-temperature cooling water and a water inlet temperature of low-temperature cooling water), a water outlet temperature of cooling water of each chiller (including a water outlet temperature of medium-temperature cooling water and a water outlet temperature of low-temperature cooling water), a primary pump frequency (including a water inlet frequency of each medium-temperature primary pump and a water inlet frequency of each low-temperature primary pump), a secondary pump frequency (including a water outlet temperature secondary pump frequency of each and a water outlet temperature secondary pump frequency of each low-temperature secondary pump frequency), a cooling pump frequency, a cooling tower frequency, an operating state, and the like. Among the above listed system parameters, the electric quantity of the ammeter, the external environment temperature, the water supply temperature of the medium/low temperature chilled water, the water outlet temperature and the water flow are non-adjustable parameters, and the others are adjustable parameters. The adjustable parameters are system parameters which need to be optimally controlled.

Of course, the above system parameters are merely exemplary, and actual system parameters include, but are not limited to, the above examples, and parameters in an actual system include hundreds. It will be appreciated by those skilled in the art that the system parameters that are changed may be adjustable parameters and may not be changed to non-adjustable parameters in the event that the system functionality requirements are met.

The adjustable parameters of various devices in the cold and heat source system are uniformly adjusted in a central control system or an automatic control system of a factory, and thus, the control method of the cold and heat source system in the embodiment of the disclosure is realized by a control device, and the control device can be realized by a computer device with data processing capability, specifically, the computer device can be a computer with data processing capability, including a personal computer (PC, personal Computer), a small computer or a large computer. The computer equipment can be specifically realized as a central control system or an automatic control system of the cold and heat source equipment. The control method of the cold and heat source system is embodied as software or a plug-in loaded in the computer device, and is executed when the software or the plug-in is loaded.

Next, a control method of the heat and cold source system provided in the present embodiment will be described from the viewpoint of a processing apparatus having data processing capability.

Prior to step S1, the control method of the embodiments of the present disclosure first trains and obtains a trained energy consumption relation model.

Alternatively, referring to fig. 3, in step S01, an energy consumption relation model is constructed, the constructed energy consumption relation model including energy consumption parameters.

Considering that the system parameters comprise hundreds of factors, by researching the influence factors of the energy consumption during the system operation and compromising the operation speed and operation cost of the model, the inventor refines the environment temperature and the total cold as key parameters, adopts a linear regression analysis model, and constructs an energy consumption relation model as follows: electric quantity=a1 ambient temperature P total cooling capacity + a2 total cooling capacity + C,

wherein, the electric quantity represents the total electric consumption of the cold and heat source system, the ambient temperature represents the temperature of the space where the cold and heat source system is located, A1, A2, P and C are energy consumption parameters, in addition, "≡P" represents the power P, the ambient temperature≡P represents the power P of the ambient temperature;

total coldness = Σ (return water temperature of each device-supply water temperature of each device) × water flow rate × fixed parameter. The total cooling capacity represents the sum of the cooling capacities of each device, wherein the part after the summation symbol represents the cooling capacity of each device, wherein the training energy consumption parameter represents an unknown parameter that needs to be determined by model training.

In step S02, a sample set for constructing an energy consumption relation model is obtained in a second preset time period, where the second preset time period is the same as or different from the first time period.

The second preset time period is a relative expression described with respect to the first preset time period. In determining the optimal control value using the control method of the present disclosure, a part of the system parameters for a first preset time period is required, and the first preset time period is a settable time period, for example, 1 year. The trained energy consumption relation model can be pre-trained and pre-stored and directly applied when the optimization control is performed; or may be trained and obtained in the time of performing the sub-optimal control or in the near term of the optimal control. If the former is, the partial system parameter for training may be a parameter of another period different from the first preset period, and if the latter is, the same system parameter may be used for training and performing the optimization control.

Based on the above consideration, the first preset time period and the second preset time period may be the same or different. For convenience of description, in this example, the first preset time period and the second preset time period are the same, that is, the obtained partial system parameters in the same time period are used for model training and used for classifying the actual electric quantity, which is not described in detail below. Of course, it is not necessary to acquire all the system parameters of the cold and heat source system, and only several parameters for constructing the energy consumption relation model and for electric quantity prediction, and adjustable parameters in the system parameters, are required.

In addition, the manner of acquiring the system parameters is not unique, and various system parameters can be acquired through communication connection between the control device and various sensors provided in various devices or environments of the system.

To obtain a varying ambient temperature, a longer time span needs to be chosen, for example 1 year, but a large number of parameters of the system within 1 year will result in a huge calculation, whereas the parameter variation within a smaller time period is not significant, for example the ambient temperature is hardly changed within half an hour.

Thus, the acquired parameters are data processed prior to constructing the sample set.

Step S02 further includes:

acquiring system parameters for constructing an energy consumption relation model in a second preset time period;

dividing the second preset time period into a plurality of time windows, for example, half an hour into one time window, and dividing the 1-year time into a plurality of time windows;

In embodiments of the present disclosure, the sampling process includes averaging, maximizing, etc., data over a time window, such as averaging temperature, frequency, flow, maximizing power.

In particular to the present example, because the system parameters used in the constructed energy consumption relationship model are the amount of electricity, the ambient temperature, and the total cooling capacity. The parameters usable here are therefore also the electric quantity, which likewise represents the total electric consumption of the cold and heat source system, the ambient temperature, which represents the temperature of the space in which the cold and heat source system is located, and the total cold, which is calculated as described in the above expression.

Optionally, before constructing the sample set, outliers are further removed, and parameters after outliers are removed are used as the sample set.

In step S03, the training set and the test set are divided based on the sample set.

The parameters of 80% in the sample set may be randomly selected as the training set and the remaining 20% as the test set, although this ratio is not limiting and other reasonable ratios are possible.

In step S04, the energy consumption relation model is trained based on the training set, and the energy consumption parameters are fitted with the fitting curve formed by the training set to obtain the energy consumption relation model to be tested.

The training process is in fact a process of continually fitting to determine unknown parameters using known true values. Referring to the figure, blue color is true value in the figure, namely, a sample set constructed by the real electric quantity, the environment temperature and the total cold quantity is obtained from the cold and heat source system, red color is a predicted value, namely, a predicted curve obtained by inputting the sample set into the energy consumption relation model for training is obtained, and the energy consumption parameters to be determined can be obtained through fitting by using a fitting curve. The larger the number of sample sets, the more accurate the energy consumption parameters obtained by training.

In step S05, the energy consumption relation model is tested with a test set to verify the trained energy consumption relation model.

Alternatively, the energy consumption relationship model is trained and tested with K-fold Cross-validation. K in the K-fold cross validation method represents that data in a sample set are randomly divided into K groups, wherein a K-1 group sample is used as a training set for training, a K group sample is used as a test set, a group of energy consumption parameters and test results (for example, the accuracy rate is 90%) are obtained after training, then the data are randomly divided into K groups from the sample set, training and testing are performed again to obtain another group of energy consumption parameters and test results, a plurality of groups of energy consumption parameters and test results are obtained after repeated for a plurality of times (less than or equal to K times), and a group of parameters with the highest accuracy rate can be selected as the energy consumption parameters of the energy consumption relation model.

Alternatively, 5K cross-validation may be selected for testing.

Of course, if the accuracy of the constructed energy consumption relation model is higher than a preset value after multiple tests, the relation of the constructed energy consumption relation model is accurately available. In the embodiment of the disclosure, after test verification, the accuracy of the energy consumption relation model constructed above is more than 90%, and the model can be used.

With continued reference to fig. 4, it can be seen that the test values (or fitted values) obtained using the trained energy consumption relationship model are mostly fitted to the true values, with only slight deviations in individual values, and also visually verifying the effect of 90% accuracy. Wherein, the abscissa is ambient temperature, cold, the ordinate is electric quantity, the blue is a true value, and the red represents a test value (or referred to as a simulated fit value).

After the trained energy consumption relation model is obtained through the training process, the trained energy consumption relation model is used for controlling the cold and heat source system.

In step S1, system parameters for power prediction are acquired.

In particular, only the ambient temperature and the total cooling capacity should be used for the trained energy consumption relation model, so that only the ambient temperature and the parameters for calculating the total cooling capacity need to be acquired.

In step S2, parameter training is performed based on the system parameters by using the trained energy consumption relation model, and a predicted electric quantity is obtained.

Specifically, in this example, the obtained system parameters are operated to obtain the total cold capacity of the cold and heat source system, and the total cold capacity and the ambient temperature are brought into a trained energy consumption relation model to obtain the predicted electric quantity.

In step S3, a first type value and a second type value in the actual electric quantity of the cold and heat source system in a first preset time period are obtained by taking the predicted electric quantity as a reference value, the value of the first type value is lower than the reference value, and the value of the second type value is higher than the reference value.

Specifically, in this step, the actual electric quantity of the cold and heat source system in the first preset time is obtained in advance, where the actual electric quantity is the total electric quantity of the system.

Optionally, the actual electric quantity of the cold and heat source system in the first preset time period is that after all system parameters in the first preset time period are acquired, a window is divided based on the ambient temperature and the total cold quantity, for example, the ambient temperature interval is [0,30], the total cold quantity interval is [50000,280000], the temperature split window is [0,5], [5,10], [10,15], [15,20], [25,30], and the total cold quantity split window is [50000,90000], [90000,130000], …, and [250000,280000]. And sampling data according to various parameters in the window, such as temperature, frequency, flow averaging, and electric quantity maximum value. Here, the actual electric quantity is the total electric quantity of the sensed cold and heat source system.

After the sampling processing, the system parameters in the window are represented by the parameters after the sampling processing.

Specifically, the electric quantity after sampling processing is divided based on a reference value, the division higher than the reference value is a first class value, and the division lower than the reference value is a second class value. For example, if the predicted electric quantity is 1000 degrees, the actual electric quantity of 950 degrees is divided into a first class value, the actual electric quantity of 1500 degrees is divided into a second class value, and so on in the actual electric quantity after sampling.

In step S4, the distribution of the adjustable parameters corresponding to the first class of values is obtained as a first distribution, and the distribution of the adjustable parameters corresponding to the second class of values is obtained as a second distribution.

Considering that only the adjustable parameters can be optimally controlled, after classification based on the actual electric quantity, only the adjustable parameters are considered, and the distribution rule of the adjustable parameters is obtained. Of course, the adjustable parameters within the first preset time period have been acquired in advance.

Optionally, for each of the adjustable parameters, a distribution of the actual electric quantity belonging to the first class of values is obtained separately as a first distribution, and a distribution of the actual electric quantity belonging to the second class of values is obtained as a second distribution. The control means acquire a probability distribution of these parameters for subsequent variance analysis.

Optionally, in order to more intuitively present the difference between the first distribution and the second distribution of each parameter, the first distribution curve is drawn based on the first distribution and the second distribution curve is drawn based on the second distribution, and the first distribution curve and the second distribution curve of the same adjustable parameter are drawn and stored in the same coordinate system as the distribution curve of the adjustable parameter. When the user intuitively verifies or analyzes, each distribution curve can be called.

Illustratively, referring to fig. 5, a probability distribution curve of 10 parameters is presented, where the ordinate represents the probability density, the abscissa represents the range of values of the parameter distribution, the blue curve represents the distribution curve of the parameter corresponding to the first class of values, and the red curve represents the distribution curve of the parameter corresponding to the second class of values, and as can be seen intuitively, the 2 nd and 4 th (respectively represented by red boxes) distribution curves in the second row show significant distribution differences. In the 2 nd curve: the red curves are substantially below 250, and the blue curves have 40% of data above 250; in curve 4: the proportion of the red curve between the intervals 200,250 is low and negligible, and the blue curve has a value of 50% in this interval.

Of course, those skilled in the art will appreciate that there are hundreds of probability distribution curves actually plotted, and only 10 are randomly selected for illustration, and the other parameters are processed in the same way.

The inventors simultaneously validated with an algorithm for nonlinear dimension reduction T-SNE (T-distributed stochastic neighbor embedding) for the adjustable parameters. Referring to fig. 6, even if the dimension reduction visualization is performed by the T-SNE algorithm that is advantageous in distinguishing the distribution of data, if the data is directly processed without the distribution classification, all the data are still mixed together, and the classification cannot be represented and the distribution rule cannot be seen. Therefore, an optimum value lower than the reference value and higher than the reference value cannot be obtained. Referring to fig. 7, several tens of adjustable parameters having different distributions are visualized by using TSNE in a dimension reduction, and it is found that red (higher than the parameter corresponding to the reference value) and blue (lower than the parameter corresponding to the reference value) can be clearly distinguished. Thereby verifying the reliability of the optimizable parameters and the optimal control values determined by the distribution differences.

Based on this, in step S5, a difference value between the first distribution and the second distribution is calculated, and when the difference value is greater than a preset threshold value, the corresponding adjustable parameter is used as the optimizable parameter.

Specifically, the JS divergence value is calculated based on the first distribution and the second distribution to automatically obtain the parameter having a significant difference among all the adjustable parameters. The JS divergence value is a difference value, and the JS divergence value calculation meets the following conditions:

wherein P (x) represents a firstThe distribution, Q (x), represents the second distribution, and operator KL (A B) represents

The probability distribution calculation method comprises the following steps: for example: the first class of values has parameters x1, x2, …, xn, the first distribution of which isThe second distribution is the same, the value range of JS divergence value calculated according to JS divergence is [0,1]. The two distributions are not different and are 0, the two distributions are completely different and are 1, and the larger the difference is, the larger the JS divergence value is. Setting a divergence threshold (such as 0.8), comparing the JS divergence value calculated by each adjustable parameter with the divergence threshold, and determining the parameters with difference in judgment higher than the divergence threshold as optimizable parameters; parameters below the divergence threshold, which are judged to be non-divergent, are not considered for optimization.

In step S6, an optimal control value for the optimizable parameter is determined based on the first distribution of the optimizable parameter.

After the control device determines the optimizable parameters through comparison, the control device determines the value of the optimization control value according to the distribution rule of the first distribution of the actual electric quantity lower than the reference value.

Optionally, after determining the optimal control value of the optimizable parameter based on the first distribution of optimizable parameters, the control method further comprises:

That is, under the same environmental temperature and total amount conditions, the point with high distribution probability in the corresponding parameter distribution is the optimal control point and the optimal set value, which is lower than the reference specification power consumption.

For further visualization, the following analysis will be described with reference to the probability distribution curve in fig. 5, and it will be understood by those skilled in the art that the basis of the analysis is a dataized distribution for the control device.

Referring to fig. 5, in the second row of the 2 nd parameters, the red curves are substantially lower than 250, that is, when the optimal control value is set to not more than 250, the energy consumption of the cold and heat source system can be lower than the reference.

Further alternatively, determining the optimal control value may also take into account a distribution of the first type of values and the second type of values. Still taking the second row of parameters 2 in fig. 5 as an example, the red curve (the parameter with the actual electric quantity lower than the reference) is basically lower than 250, and the blue curve (the parameter with the actual electric quantity higher than the reference) has a value of 40% higher than 250, it can be determined that the set value higher than 250 has a low probability of obtaining the energy consumption lower than the reference, and correspondingly, a relatively increased probability of obtaining the energy consumption higher than the reference. Thus, comprehensively considering, the parameter is set to be lower than 250 as the optimal control value.

In other words, the optimal control scheme is set when the optimal control value is set to a region in the first distribution in which the distribution probability is higher while the distribution probability is lower, based on the first distribution and the second distribution of the optimizable parameter.

It will be appreciated by those skilled in the art that controlling the cold heat source system with the determined optimal control value as the updated control value enables the system to operate at an optimal level by adjusting each optimizable parameter to an optimal control value that enables the system to operate at a low energy consumption level.

Through the arrangement, in the embodiment, the predicted electric quantity is obtained through the trained energy consumption relation model, the predicted electric quantity is used as a reference, the actual electric quantity is divided into the first class value and the second class value, the distribution of each adjustable parameter is obtained aiming at the first class value and the second class value, the parameters with large distribution difference are redistributed, and the optimal control value is determined according to the parameter distribution rule lower than the reference value in the extracted parameters with large distribution difference, so that the control value of the whole system can be optimized, the cold and hot source system is controlled at a low energy consumption level, and the energy consumption of the system is reduced.

Based on the same inventive concept, as shown in fig. 8, an embodiment of the present disclosure further provides a control device of a cold and heat source system, which is characterized by comprising:

An acquisition unit 101 that acquires system parameters for the energy consumption relation model;

the prediction unit 102 obtains a predicted electric quantity by using the trained energy consumption relation model based on the system parameters for the energy consumption relation model;

the classifying unit 103 takes the predicted electric quantity as a reference value, and obtains a first class value and a second class value in the actual electric quantity of the cold and heat source system in a first preset time period, wherein the value of the first class value is lower than the reference value, and the value of the second class value is higher than the reference value;

the analysis unit 104 obtains the distribution of the adjustable parameters corresponding to the first class value as a first distribution, and obtains the distribution of the adjustable parameters corresponding to the second class value as a second distribution;

the calculating unit 105 calculates a difference value between the first distribution and the second distribution, and when the difference value is greater than a preset threshold value, the corresponding adjustable parameter is used as an optimizable parameter;

the optimizing unit 106 determines an optimized control value of the optimizable parameter based on the first distribution of the optimizable parameter.

It should be understood by those skilled in the art that the control device of the cold-heat source system includes the above units, but is not limited to include only the above units, and in some embodiments the control device of the cold-heat source system may include an interface for outputting a probability distribution curve, that is, further includes an output unit, and specific functions have been described in detail in the embodiments describing the control method of the cold-heat source system, which is not described herein.

In this embodiment, the control device predicts the predicted electric quantity through the trained energy consumption relation model, divides the actual electric quantity into a first class value and a second class value based on the predicted electric quantity, obtains the distribution of each adjustable parameter according to the first class value and the second class value, redistributes the parameters with large differences, and determines the optimized control value according to the parameter distribution rule lower than the reference value in the extracted parameters with large distribution differences, thereby optimizing the control value of the whole system, controlling the cold and heat source system at a low energy consumption level, reducing the energy consumption of the system, and having wide application prospects.

Another embodiment of the present disclosure provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements:

s1, acquiring system parameters for electric quantity prediction;

s2, acquiring predicted electric quantity by using a trained energy consumption relation model based on system parameters for electric quantity prediction;

In practical applications, the computer-readable storage medium may take the form of any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this embodiment, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

The computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

As shown in fig. 9, another embodiment of the present disclosure provides a schematic structural diagram of a computer device. The computer device 12 shown in fig. 9 is merely an example and should not be construed as limiting the functionality and scope of use of the disclosed embodiments.

As shown in fig. 9, the computer device 12 is in the form of a general purpose computing device. Components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, a bus 18 that connects the various system components, including the system memory 28 and the processing units 16.

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, and a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

The system memory 28 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 30 and/or cache memory 32. The computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from or write to non-removable, nonvolatile magnetic media (not shown in FIG. 9, commonly referred to as a "hard disk drive"). Although not shown in fig. 9, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be coupled to bus 18 through one or more data medium interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of the various embodiments of the disclosure.

A program/utility 40 having a set (at least one) of program modules 42 may be stored in, for example, memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods in the embodiments described in this disclosure.

The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the computer device 12, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 12 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 22. Moreover, computer device 12 may also communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through network adapter 20. As shown in fig. 9, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be appreciated that although not shown in fig. 9, other hardware and/or software modules may be used in connection with computer device 12, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.

The processor unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, for example, implementing an automatic demonstration method of an android debug bridge-based application program applied to a terminal installed with an application program provided in an embodiment of the present disclosure, or implementing a control method of a cold and heat source system provided in an embodiment of the present disclosure.

Aiming at the existing problems at present, the control method and device, the computer readable storage medium and the computer equipment of the cold and heat source system are formulated, the energy consumption higher than the predicted electric quantity and the energy consumption higher than the predicted electric quantity are divided by taking the predicted electric quantity as a reference, the adjustable parameters with larger difference in distribution are determined for the two types of energy consumption, and the optimal control value is determined based on the probability distribution, so that the adjustable parameters with larger influence on the actual electric quantity can be determined, accurate parameter optimization is performed, the energy consumption of the system is reduced, the energy saving of the system is realized, and the system has wide application prospect.

It should be apparent that the foregoing examples of the present disclosure are merely illustrative of the present disclosure and not limiting of the embodiments of the present disclosure, and that various other changes and modifications may be made by one of ordinary skill in the art based on the foregoing description, and it is not intended to be exhaustive of all embodiments, and all obvious changes and modifications that come within the scope of the present disclosure are intended to be embraced by the technical solution of the present disclosure.

Claims

1. A control method of a cold and heat source system, comprising:

Acquiring system parameters for electric quantity prediction;

acquiring predicted electric quantity by using a trained energy consumption relation model based on the system parameters for electric quantity prediction;

2. The control method of claim 1, wherein the determining an optimal control value for the optimizable parameter based on the first distribution of the optimizable parameter further comprises:

And the distribution interval is used as a value interval of the optimized control value to determine the optimized control value of the optimized parameter.

3. The control method according to claim 1, characterized in that before the predicted electric quantity is obtained using the trained energy consumption relation model based on the system parameters for electric quantity prediction, the method further comprises:

dividing a training set and a testing set based on the sample set;

training the energy consumption relation model based on the training set, and fitting the energy consumption parameters by using a fitting curve formed by the training set to obtain an energy consumption relation model to be tested;

and testing the energy consumption relation model by the test set to verify the trained energy consumption relation model.

4. A control method according to claim 3, characterized in that the constructed energy consumption relation model is: electric quantity=a1 ambient temperature P total cooling capacity + a2 total cooling capacity + C,

Wherein the electric quantity represents the total electric consumption of the cold and heat source system, the ambient temperature represents the temperature of the space where the cold and heat source system is positioned, A1, A2, P and C are the energy consumption parameters,

5. The control method according to claim 1, characterized in that the calculating a difference value between the first distribution and the second distribution further comprises:

and calculating a JS divergence value based on the first distribution and the second distribution, wherein the JS divergence value is the difference value, and the calculation of the JS divergence value meets the following conditions:

6. The control method according to claim 1, wherein the adjustable parameter includes: one or more of a chilled water inlet temperature of the chiller, a chilled water outlet temperature of the chiller, a primary pump frequency, a secondary pump frequency, a cooling tower frequency, and an operational status.

7. The control method according to claim 3, wherein the acquiring the sample set based on the system parameters for constructing the energy consumption relation model during the second preset period of time further comprises:

dividing the second preset time period into a plurality of time windows;

sampling the system parameters for constructing the energy consumption relation model in the second preset time period by taking the time window as a time unit to construct the sample set.

8. The control method according to claim 1, wherein after the obtaining the distribution of the adjustable parameter corresponding to the first class value as a first distribution and the obtaining the distribution of the adjustable parameter corresponding to the second class value as a second distribution, the control method further comprises:

drawing a first distribution curve based on the first distribution and drawing a second distribution curve based on the second distribution,

and drawing the first distribution curve and the second distribution curve of the same adjustable parameter as the distribution curve of the adjustable parameter in the same coordinate system and storing the first distribution curve and the second distribution curve.

9. A control device for a cold and heat source system, comprising:

the prediction unit is used for obtaining predicted electric quantity by utilizing the trained energy consumption relation model based on the system parameters for the energy consumption relation model;

The classification unit takes the predicted electric quantity as a reference value, and obtains a first class value and a second class value in the actual electric quantity of the cold and heat source system in a first preset time period, wherein the value of the first class value is lower than the reference value, and the value of the second class value is higher than the reference value;

10. A computer-readable storage medium having a computer program stored thereon, characterized in that,

the program, when executed by a processor, realizes the control method of the cold heat source system according to any one of claims 1 to 8.

11. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that,

The control method of the cold and heat source system according to any one of claims 1 to 8 is realized when the processor executes the program.