CN110542178B - Air conditioner and refrigerator room control method and system with self-learning capability - Google Patents

Abstract

The invention aims to provide an air conditioner refrigeration machine room control method and system with self-learning capability.

Description

Air conditioner and refrigerator room control method and system with self-learning capability

Technical Field

The invention relates to the field of computers, in particular to a control method and a system for an air-conditioning and refrigerating machine room with self-learning capability.

Background

The traditional machine room group control system is controlled by a standardized program of single equipment, the overall efficiency and energy conservation of a machine room are not considered, and an overall data model of the machine room system is lacked; and the traditional algorithm can not carry out quantitative modeling analysis on the coupling relation of a plurality of environment variables such as temperature, pressure, flow and the like, so that an energy-saving model of the whole system of the machine room can not be found, the group control system only carries out basic operations such as sequential start-stop and the like on main equipment such as a cold machine, a water pump, a cooling tower and the like, the parameters of the machine room can not be finely adjusted, and the energy-saving effect is often not ideal.

Disclosure of Invention

The invention aims to provide an air conditioner and refrigerating machine room control method and system with self-learning capability.

According to one aspect of the invention, an air conditioner and refrigeration machine room control method with self-learning capability is provided, and comprises the following steps:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises temperature, flow, pressure and current data of a host machine, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01 and R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

According to another aspect of the present invention, there is also provided an air conditioning and refrigerating machine room control system with self-learning capability, wherein the system comprises:

the system comprises a first device, a second device and a third device, wherein the first device is used for collecting operation condition parameter groups of equipment of an air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, and each group of operation condition parameter groups comprise temperature, flow, pressure and current data of a host, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

the second device is used for dividing the operating condition parameter group in the database into two parts, wherein one part is used as a training parameter set, and the other part is used as a verification parameter set;

a third device, configured to train based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, where a formula of the COP efficiency prediction model is as follows:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

fourth means for evaluating the COP efficiency prediction model based on the following equation:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

a fifth means for repeatedly adjusting w in the COP efficiency model if MSE is 0.01 or more and R is 0.1 or more_iAnd b until MSE is less than 0.01 and R is less than 0.1;

a sixth device, configured to use the training parameter set and the verification parameter set as a set Q, find a corresponding operating condition parameter set when COP efficiency is optimal from the set Q, download the operating condition parameter set as an input parameter at time t to a relevant execution mechanism of an air-conditioning refrigeration room, execute the parameter, and obtain a corresponding actual value of COP efficiency after the execution of the input parameter at time t;

a seventh device, configured to use an O-U process to search and obtain multiple sets of randomly searched operating condition parameter sets;

the eighth device is used for calculating COP efficiency predicted values of all groups of randomly explored operation condition parameter sets through the COP efficiency prediction model;

a ninth device, configured to obtain, from the sets of randomly explored operating condition parameter sets, a set of randomly explored operating condition parameter sets corresponding to a set having a highest COP efficiency prediction value, as an operating condition parameter set a;

the tenth device is used for downloading the group A of the operating condition parameters as input parameters at the time of t +1 to a related actuating mechanism of the air-conditioning refrigeration machine room for execution and then acquiring a corresponding COP (coefficient of performance) actual value after the execution of the input parameters at the time of t + 1;

eleventh means for comparing the actual value of the COP efficiency at time t with the actual value of the COP efficiency at time t +1,

a twelfth means for determining whether the set Q has the operating condition parameter group a if the COP efficiency actual value at the time t +1 is greater than or equal to the COP efficiency actual value at the time t, adding the operating condition parameter group a to the set Q if the set Q does not have the operating condition parameter group a, and increasing the selected probability value of the operating condition parameter group a in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

a thirteenth device, configured to determine whether the set Q has the operating condition parameter set a if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, add the operating condition parameter set a to the set Q if the set Q does not have the operating condition parameter set a, and reduce a selected probability value of the operating condition parameter set a in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

fourteenth means for cyclically and sequentially executing the seventh to fourteenth means after the time t +1 is taken as a new time t until the cycle exceeds a preset time threshold;

and the fifteenth device is used for acquiring the corresponding operation condition parameter group B with the highest selected probability value from the set Q after the cycle exceeds the preset time threshold value, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigerating machine room for execution.

According to another aspect of the present invention, there is also provided a computing-based device, including:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises temperature, flow, pressure and current data of a host machine, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01 and R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

According to another aspect of the present invention, there is also provided a computer-readable storage medium having stored thereon computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, cause the processor to:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises temperature, flow, pressure and current data of a host machine, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is greater than or equal to 0.01 and R is greater than or equal to 0.1, repeatedly adjusting the COP efficiencyW in the model_iAnd b until MSE is less than 0.01 and R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

Compared with the prior art, the method can self-learn and dynamically adjust the input parameters according to the on-site actual working conditions of the air-conditioning refrigeration machine room by the COP efficiency prediction model and the selected probability value of the operation working condition parameter set, find the operation working condition parameter set corresponding to the maximization of the integral COP efficiency of the system, and achieve the energy-saving effect better than that achieved by the traditional control mode of the chiller group control system.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:

fig. 1 shows a flow chart of a control method of an air conditioner/freezer room with self-learning capability according to an embodiment of the present invention.

The same or similar reference numbers in the drawings identify the same or similar elements.

Detailed Description

The present invention is described in further detail below with reference to the attached drawing figures.

In a typical configuration of the present application, the terminal, the device serving the network, and the trusted party each include one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.

As shown in fig. 1, the present invention provides a control method for an air conditioning and refrigerating machine room with self-learning capability, which comprises the following steps:

step S01, collecting the operation condition parameters of the equipment of the air-conditioning refrigeration machine room and putting the parameters into a database;

step S02, dividing the operation condition parameters in the database into two parts, one part being used as a training parameter set and the other part being used as a verification parameter set;

step S03, training based on the training parameter set through a preset neural network algorithm to obtain a COP (coefficient of performance/input power) efficiency prediction model;

and step S04, obtaining corresponding operation condition parameters when the COP efficiency is the highest based on the COP efficiency prediction model.

The invention provides a control method of an air-conditioning and refrigerating machine room with self-learning capability, which comprises the following steps:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises temperature, flow, pressure and current data of a host machine, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

all the operation condition parameter sets can be sorted and screened, the operation condition parameter sets under the stable condition are intercepted and stored, and the operation condition parameter sets collected under the unstable condition are deleted;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP (cooling capacity/input power) efficiency prediction model, wherein a formula of the COP efficiency prediction model is as follows:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

here, ANN represents a model-free black-box computational strategy, which consists of interconnected artificial neurons; each with input/output (I/O) characteristics and implementing local calculations, the COP efficiency prediction model has r input artificial neurons, a weight w is assigned to each input parameter u to describe its influence (strength), the sum of the weighted inputs and the bias b forms the input of an activation function f, which may be linear or non-linear differentiable, then gives the output a of the neuron.

Step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

herein, neural networks may be described as machine learning techniques. Modifying the values of the connection weights of the neural networks by a certain training algorithm to enable the network to be close to the solution of a system model, wherein the learning capability of the neural networks depends on the optional selection of the architecture and the training algorithm, and the selection of the activation function can obviously influence the applicability of the training algorithm;

in this process, the performance of the network can be evaluated by Mean Square Error (MSE), for any output variable in the simulation process, whose correlation coefficient R (compare the complete set of actual outputs with the target) is indicative of the extent to which the actual outputs can explain the target, if the R values of all output variables are small, such as less than 0.1, the COP efficiency model can be considered as a good representation of the system;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01 and R is less than 0.1;

the unit system of the air-conditioning refrigeration machine room is normalized under the wide-range load condition of 30-100%, the numerical value of test data is normalized to be in the range of-1, and a better training result is obtained after final training iteration, wherein the MSE is less than 0.01, and the R value is less than 0.1, which means that the prediction error of the system COP under a relevant model is not more than 10%;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

in the process of searching for the optimal COP in a certain period, random exploration of different input variables can be continuously added, and an O-U process (Ornstein-Uhlenbeck process) is used as a random process of exploration;

step S8, predicting COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

Here, the time t is the previous time, the time t +1 is the next time after the time t, and when the time t +1 is the time t, the COP efficiency actual value at the time t +1 is also changed to the COP efficiency actual value at the time t, and the COP efficiency actual value at the next time t +1 can be cyclically acquired.

The training model of a water chiller group control system is researched by adopting the artificial neural network algorithm. The neural network algorithm used trained a model with a 3-tier network, 4 input variables and 1 output variable. The model is used to find the optimal combination of chilled water and cooling water system settings in order to minimize the total energy cost of the system at various cooling loads, i.e. to optimize the overall COP (cooling capacity/input power) efficiency of the system. Thus, the input includes 4 independent control variables (sets of operating condition parameters): the set temperature of chilled water supply (T1), the chilled water and cooling water flow (m1, m2), the cooling water return temperature (T4) and the refrigerating capacity (Qc) of the water chiller. The output variable (COP) accurately reflects the efficiency of the chiller system.

According to the method, the COP efficiency prediction model and the selected probability value of the operation condition parameter set are obtained, the input parameters can be dynamically adjusted in a self-learning manner according to the field actual conditions of the air-conditioning refrigeration machine room, the operation condition parameter set corresponding to the system overall COP efficiency maximization is searched, and the energy-saving effect better than that achieved by the traditional refrigerator group control system control mode is achieved.

According to another aspect of the present invention, there is also provided an air conditioning and refrigerating machine room control system with self-learning capability, wherein the system comprises:

the system comprises a first device, a second device and a third device, wherein the first device is used for collecting operation condition parameter groups of equipment of an air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, and each group of operation condition parameter groups comprise temperature, flow, pressure and current data of a host, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

the second device is used for dividing the operating condition parameter group in the database into two parts, wherein one part is used as a training parameter set, and the other part is used as a verification parameter set;

a third device, configured to train based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, where a formula of the COP efficiency prediction model is as follows:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

fourth means for evaluating the COP efficiency prediction model based on the following equation:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iPresentation verification parameterActual values of COP efficiency of the ith group of operating condition parameter sets in the data sets,

an absolute value representing an average of the n COP efficiency predictions;

a fifth means for repeatedly adjusting w in the COP efficiency model if MSE is 0.01 or more and R is 0.1 or more_iAnd b until MSE is less than 0.01 and R is less than 0.1;

a sixth device, configured to use the training parameter set and the verification parameter set as a set Q, find a corresponding operating condition parameter set when COP efficiency is optimal from the set Q, download the operating condition parameter set as an input parameter at time t to a relevant execution mechanism of an air-conditioning refrigeration room, execute the parameter, and obtain a corresponding actual value of COP efficiency after the execution of the input parameter at time t;

a seventh device, configured to use an O-U process to search and obtain multiple sets of randomly searched operating condition parameter sets;

the eighth device is used for calculating COP efficiency predicted values of all groups of randomly explored operation condition parameter sets through the COP efficiency prediction model;

a ninth device, configured to obtain, from the sets of randomly explored operating condition parameter sets, a set of randomly explored operating condition parameter sets corresponding to a set having a highest COP efficiency prediction value, as an operating condition parameter set a;

the tenth device is used for downloading the group A of the operating condition parameters as input parameters at the time of t +1 to a related actuating mechanism of the air-conditioning refrigeration machine room for execution and then acquiring a corresponding COP (coefficient of performance) actual value after the execution of the input parameters at the time of t + 1;

eleventh means for comparing the actual value of the COP efficiency at time t with the actual value of the COP efficiency at time t +1,

a twelfth means for determining whether the set Q has the operating condition parameter group a if the COP efficiency actual value at the time t +1 is greater than or equal to the COP efficiency actual value at the time t, adding the operating condition parameter group a to the set Q if the set Q does not have the operating condition parameter group a, and increasing the selected probability value of the operating condition parameter group a in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

a thirteenth device, configured to determine whether the set Q has the operating condition parameter set a if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, add the operating condition parameter set a to the set Q if the set Q does not have the operating condition parameter set a, and reduce a selected probability value of the operating condition parameter set a in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

fourteenth means for cyclically and sequentially executing the seventh to fourteenth means after the time t +1 is taken as a new time t until the cycle exceeds a preset time threshold;

and the fifteenth device is used for acquiring the corresponding operation condition parameter group B with the highest selected probability value from the set Q after the cycle exceeds the preset time threshold value, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigerating machine room for execution.

According to another aspect of the present invention, there is also provided a computing-based device, including:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises temperature, flow, pressure and current data of a host machine, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01 and R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

According to another aspect of the present invention, there is also provided a computer-readable storage medium having stored thereon computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, cause the processor to:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises temperature, flow, pressure and current data of a host machine, a water pump, a cooling tower, an electric valve and a sensor of the air-conditioning and refrigerating machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01 and R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

For details of embodiments of each device and storage medium of the present invention, reference may be made to corresponding parts of each method embodiment, and details are not described herein again.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

It should be noted that the present invention may be implemented in software and/or in a combination of software and hardware, for example, as an Application Specific Integrated Circuit (ASIC), a general purpose computer or any other similar hardware device. In one embodiment, the software program of the present invention may be executed by a processor to implement the steps or functions described above. Also, the software programs (including associated data structures) of the present invention can be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Further, some of the steps or functions of the present invention may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.

In addition, some of the present invention can be applied as a computer program product, such as computer program instructions, which when executed by a computer, can invoke or provide the method and/or technical solution according to the present invention through the operation of the computer. Program instructions which invoke the methods of the present invention may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the invention herein comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or solution according to embodiments of the invention as described above.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims (4)

1. A control method for an air conditioning and refrigerating machine room with self-learning capability comprises the following steps:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises: temperature, flow, pressure and current data of a sensor of a host of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a water pump of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a cooling tower of the air-conditioning refrigeration machine room, and temperature, flow, pressure and current data of a sensor of an electric valve of the air-conditioning refrigeration machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01, R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

2. An air conditioning and refrigeration machine room control system with self-learning capability, wherein the system comprises:

the first device is used for collecting operation condition parameter groups of equipment of an air-conditioning refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprise: temperature, flow, pressure and current data of a sensor of a host of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a water pump of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a cooling tower of the air-conditioning refrigeration machine room, and temperature, flow, pressure and current data of a sensor of an electric valve of the air-conditioning refrigeration machine room;

the second device is used for dividing the operating condition parameter group in the database into two parts, wherein one part is used as a training parameter set, and the other part is used as a verification parameter set;

a third device, configured to train based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, where a formula of the COP efficiency prediction model is as follows:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

fourth means for evaluating the COP efficiency prediction model based on the following equation:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

fifth means for R is greater than or equal to 0.01 if MSE is greater than or equal toEqual to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01, R is less than 0.1;

a sixth device, configured to use the training parameter set and the verification parameter set as a set Q, find a corresponding operating condition parameter set when COP efficiency is optimal from the set Q, download the operating condition parameter set as an input parameter at time t to a relevant execution mechanism of an air-conditioning refrigeration room, execute the parameter, and obtain a corresponding actual value of COP efficiency after the execution of the input parameter at time t;

a seventh device, configured to use an O-U process to search and obtain multiple sets of randomly searched operating condition parameter sets;

the eighth device is used for calculating COP efficiency predicted values of all groups of randomly explored operation condition parameter sets through the COP efficiency prediction model;

a ninth device, configured to obtain, from the sets of randomly explored operating condition parameter sets, a set of randomly explored operating condition parameter sets corresponding to a set having a highest COP efficiency prediction value, as an operating condition parameter set a;

the tenth device is used for downloading the group A of the operating condition parameters as input parameters at the time of t +1 to a related actuating mechanism of the air-conditioning refrigeration machine room for execution and then acquiring a corresponding COP (coefficient of performance) actual value after the execution of the input parameters at the time of t + 1;

eleventh means for comparing the actual value of the COP efficiency at time t with the actual value of the COP efficiency at time t +1,

a twelfth means for determining whether the set Q has the operating condition parameter group a if the COP efficiency actual value at the time t +1 is greater than or equal to the COP efficiency actual value at the time t, adding the operating condition parameter group a to the set Q if the set Q does not have the operating condition parameter group a, and increasing the selected probability value of the operating condition parameter group a in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

a thirteenth device, configured to determine whether the set Q has the operating condition parameter set a if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, add the operating condition parameter set a to the set Q if the set Q does not have the operating condition parameter set a, and reduce a selected probability value of the operating condition parameter set a in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

fourteenth means for cyclically and sequentially executing the seventh to fourteenth means after the time t +1 is taken as a new time t until the cycle exceeds a preset time threshold;

and the fifteenth device is used for acquiring the corresponding operation condition parameter group B with the highest selected probability value from the set Q after the cycle exceeds the preset time threshold value, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigerating machine room for execution.

3. A computing-based device, comprising:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises: temperature, flow, pressure and current data of a sensor of a host of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a water pump of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a cooling tower of the air-conditioning refrigeration machine room, and temperature, flow, pressure and current data of a sensor of an electric valve of the air-conditioning refrigeration machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01, R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

4. A computer-readable storage medium having computer-executable instructions stored thereon, wherein the computer-executable instructions, when executed by a processor, cause the processor to:

step S1, collecting operation condition parameter groups of equipment of the air-conditioning and refrigerating machine room and putting the operation condition parameter groups into a database, wherein each group of operation condition parameter groups comprises: temperature, flow, pressure and current data of a sensor of a host of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a water pump of the air-conditioning refrigeration machine room, temperature, flow, pressure and current data of a sensor of a cooling tower of the air-conditioning refrigeration machine room, and temperature, flow, pressure and current data of a sensor of an electric valve of the air-conditioning refrigeration machine room;

step S2, dividing the operation condition parameter group in the database into two parts, one part is used as a training parameter set, and the other part is used as a verification parameter set;

step S3, training based on the training parameter set through a preset neural network algorithm to obtain a COP efficiency prediction model, wherein the COP efficiency prediction model has the following formula:

wherein a represents COP efficiency, r represents the number of groups of operating condition parameter groups in the training parameter set, and u represents_iRepresenting the i-th set of operating condition parameters, w, in the training parameter set_iRepresenting the weight corresponding to the ith group of operating condition parameter sets in the training parameter set, and b representing the offset;

step S4, evaluating the COP efficiency prediction model based on the following formula:

wherein n represents the group number of the operating condition parameter group in the verification parameter set, a_iRepresenting a COP efficiency predicted value t calculated by the COP efficiency prediction model by the ith group of operating condition parameter sets in the verification parameter set_iRepresenting the actual value of COP efficiency of the ith group of operating condition parameter sets in the verification parameter set,

an absolute value representing an average of the n COP efficiency predictions;

step S5, if MSE is larger than or equal to 0.01 and R is larger than or equal to 0.1, repeatedly adjusting w in the COP efficiency model_iAnd b until MSE is less than 0.01, R is less than 0.1;

step S6, using the training parameter set and the verification parameter set as a set Q, searching a corresponding operation condition parameter set when the COP efficiency is optimal from the set Q, downloading the operation condition parameter set as an input parameter at the time t to a related execution mechanism of an air-conditioning refrigeration machine room for execution, and then obtaining a corresponding COP efficiency actual value after the execution of the input parameter at the time t;

step S7, acquiring a plurality of groups of randomly searched operation condition parameter sets by using O-U process search;

step S8, calculating COP efficiency prediction values of each group of randomly searched operation condition parameter groups through the COP efficiency prediction model;

step S9, obtaining a group of randomly explored operation condition parameter groups corresponding to the highest COP efficiency predicted value from the group of randomly explored operation condition parameter groups as operation condition parameter group A;

step S10, downloading the group A of the operating condition parameter sets as input parameters at the time of t +1 to relevant actuating mechanisms of an air-conditioning refrigeration machine room for execution, and acquiring corresponding COP efficiency actual values after the input parameters at the time of t +1 are executed;

step S11, comparing the COP efficiency actual value at the time t with the COP efficiency actual value at the time t +1,

step S12, if the COP efficiency actual value at the time t +1 is larger than or equal to the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and increasing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, increasing the selected probability value of the operating condition parameter group A in the set Q;

step S13, if the COP efficiency actual value at the time t +1 is smaller than the COP efficiency actual value at the time t, judging whether the set Q has the operating condition parameter group A, if the set Q does not have the operating condition parameter group A, adding the operating condition parameter group A into the set Q, and reducing the selected probability value of the operating condition parameter group A in the set Q; if the set Q has the operating condition parameter group A, reducing the selected probability value of the operating condition parameter group A in the set Q;

step S14, after the time t +1 is taken as a new time t, the step S7-step S14 are circularly and sequentially executed until the circulation exceeds a preset time threshold;

and step S15, acquiring the corresponding operation condition parameter group B with the highest probability value from the set Q after the circulation exceeds the preset time threshold, and downloading the operation condition parameter group B as an input parameter to a related execution mechanism of the air-conditioning refrigeration machine room for execution.

Images