Refrigeration plant energy efficiency optimization method, system and terminal

Abstract

The application relates to a refrigeration plant room energy efficiency optimization method and system and a terminal, and relates to the technical field of a refrigeration plant room.The application comprises the following steps: acquiring a basic predicted load and user equipment parameters; analyzing the user equipment parameters to determine a corrected predicted load; analyzing the basic predicted load and the corrected predicted load to determine a final predicted load; constructing an equipment working condition combination two-dimensional energy efficiency matrix based on the final predicted load; optimizing the equipment working condition combination two-dimensional energy efficiency matrix according to a preset PSO-GA fusion algorithm to determine a recommended operation combination; the PSO-GA fusion algorithm comprises a PSO algorithm and a GA algorithm; and controlling the refrigeration plant room to adjust an operation state according to the recommended operation combination.The application has the effects of improving the accuracy of load prediction and improving the control effect of the refrigeration plant room.

Description

Technical Field

[0001] This application relates to the technical field of refrigeration rooms, and in particular to a method, system and terminal for optimizing the energy efficiency of refrigeration rooms. Background Technology

[0002] The refrigeration room is the core cold source supply unit of the building's central air conditioning system, and its energy efficiency directly affects the building's energy consumption.

[0003] In related technologies, a chiller room typically includes a chiller unit, a cooling tower, a distribution pump set, and air conditioning terminals. The chiller unit generates low-temperature chilled water, while the cooling tower discharges the heat generated by the chiller unit during the cooling process into the atmosphere. The distribution pump set delivers the low-temperature chilled water to the air conditioning terminal equipment, absorbs indoor heat, and then returns it to the chiller unit for further cooling. The chiller room typically uses a linear prediction model built based on historical load data and outdoor temperature. The number of units and the speed of the pumps are preset according to the predicted load. The pump frequency is adjusted by feedback of the chilled water supply and return temperature difference, so that the chiller room can adjust in a timely manner according to the outdoor temperature.

[0004] Regarding the aforementioned technologies, load forecasting relies solely on historical loads and outdoor temperatures, neglecting disturbances such as human activity and the start-up of temporary equipment. For example, the impact of opening windows or starting electronic devices on the load directly results in low accuracy of load forecasting, leading to poor control performance in the chiller room, and there is still room for improvement. Summary of the Invention

[0005] To improve the accuracy of load forecasting and thus enhance the control effect of the chiller room, this application provides a method, system, and terminal for optimizing the energy efficiency of the chiller room.

[0006] Firstly, this application provides a method for optimizing the energy efficiency of a chiller room, employing the following technical solution:

[0007] A method for optimizing energy efficiency in a refrigeration room includes:

[0008] Obtain basic forecast load and user equipment parameters;

[0009] Analyze user equipment parameters to determine the revised forecast load;

[0010] The basic forecast load and the revised forecast load are analyzed to determine the final forecast load;

[0011] A two-dimensional energy efficiency matrix of equipment operating conditions is constructed based on the final predicted load.

[0012] The two-dimensional energy efficiency matrix of equipment operating condition combinations is optimized according to the preset PSO-GA fusion algorithm to determine the recommended operating combination; the PSO-GA fusion algorithm includes the PSO algorithm and the GA algorithm.

[0013] The operating status of the refrigeration room is adjusted according to the recommended operating combination.

[0014] Optionally, the step of analyzing user equipment parameters to determine the revised forecast load includes:

[0015] The user equipment parameters are calculated based on a preset independent behavior algorithm to determine the independent behavior correction load; the user equipment parameters include the user group number, the corresponding behavior type number, and the corresponding behavior intensity variable;

[0016] The user device parameters are calculated based on a preset interaction behavior algorithm to determine the interaction behavior correction load;

[0017] The modified loads for independent behavior and the modified loads for interactive behavior are analyzed to determine the modified forecast loads.

[0018] Optionally, the expression for the independent behavior algorithm is:

[0019] ,

[0020] In the formula, Adjust the load for independent behavior. The total number of user group categories. For user group serial numbers, The total number of user behavior types For behavior type sequence number, The linear influence coefficient is... For behavioral intensity variables, This is a nonlinear influence coefficient. For the maximum lag time, For the time lag, This is the hysteresis attenuation coefficient.

[0021] Optionally, the expression for the interaction behavior algorithm is:

[0022] ,

[0023] In the formula, Adjust the load for interactive behavior. For another behavior type sequence number, This represents the coefficient of influence of behavioral interactions.

[0024] Optionally, the steps for optimizing the two-dimensional energy efficiency matrix of equipment operating condition combinations based on a preset PSO-GA fusion algorithm to determine the recommended operating combinations include:

[0025] Obtain the energy consumption cost objective function;

[0026] The two-dimensional energy efficiency matrix of equipment operating condition combinations is calculated based on the PSO algorithm and the energy consumption cost objective function to determine the high-quality equipment operating condition particles.

[0027] The high-quality equipment condition particles are analyzed using the GA algorithm to determine the cross-variant equipment condition particles.

[0028] The optimal equipment condition particles are determined based on the high-quality equipment condition particles and the cross-variant equipment condition particles.

[0029] The PSO-GA fusion algorithm is used to further optimize the equipment operating condition particles to determine the recommended operating combination.

[0030] Optionally, the step of calculating the two-dimensional energy efficiency matrix of equipment operating condition combinations based on the PSO algorithm and the energy consumption cost objective function to determine the high-quality equipment operating condition particles includes:

[0031] The two-dimensional energy efficiency matrix of equipment operating condition combination is calculated based on the energy consumption cost objective function to determine the particle fitness of equipment operating conditions.

[0032] The two-dimensional energy efficiency matrix of equipment operating condition combinations and the corresponding fitness of equipment operating condition particles are analyzed to determine the basic equipment operating condition particles.

[0033] Determine whether the working conditions of the basic equipment particles meet the requirements of the preset constraints;

[0034] If it does not meet the requirements, the basic equipment condition particles will be removed.

[0035] If the conditions are met, the basic equipment operating condition particles are analyzed to determine the high-quality equipment operating condition particles.

[0036] Optionally, the step of analyzing high-quality equipment condition particles according to the GA algorithm to determine crossover variant equipment condition particles includes:

[0037] Analyze high-quality equipment operating condition particles to determine the selection of equipment operating condition particles;

[0038] Cross the selected equipment condition particles to generate cross equipment condition particles;

[0039] Mutate the cross-equipment condition particles to generate cross-mutated equipment condition particles.

[0040] Secondly, this application provides a chiller room energy efficiency optimization system, which adopts the following technical solution:

[0041] A chiller room energy efficiency optimization system, comprising:

[0042] The acquisition module is used to acquire basic forecast load and user equipment parameters;

[0043] A memory for storing a program for a method of optimizing energy efficiency in a refrigeration room as described in any of the preceding claims;

[0044] The processor and the program in the memory can be loaded and executed by the processor to implement a cooling room energy efficiency optimization method as described in any of the above.

[0045] Thirdly, this application provides a smart terminal, which adopts the following technical solution:

[0046] A smart terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any of the preceding claims, a method for optimizing energy efficiency in a refrigeration room.

[0047] In summary, this application includes at least one of the following beneficial technical effects:

[0048] 1. After analyzing the user equipment parameters, a revised predicted load is obtained. The basic predicted load is then revised using the revised predicted load to obtain the final predicted load. A two-dimensional energy efficiency matrix of equipment operating conditions is constructed based on the accurate final predicted load. The two-dimensional energy efficiency matrix of equipment operating conditions is then optimized using the PSO-GA fusion algorithm to find the recommended operating combination with the lowest energy consumption and cost that meets the final predicted load. The recommended operating combination is used to control the chiller room to adjust its operating status, thereby improving the control effect of the chiller room. Attached Figure Description

[0049] Figure 1 This is a flowchart of a method for optimizing the energy efficiency of a refrigeration room according to an embodiment of this application.

[0050] Figure 2 This is a flowchart illustrating the steps in this application embodiment to analyze user equipment parameters to determine the corrected predicted load.

[0051] Figure 3 This is a flowchart illustrating the steps in this application embodiment to optimize the two-dimensional energy efficiency matrix of equipment operating condition combinations based on a preset PSO-GA fusion algorithm to determine the recommended operating combination.

[0052] Figure 4 This is a flowchart illustrating the steps in this application embodiment to calculate the two-dimensional energy efficiency matrix of equipment operating condition combinations based on the PSO algorithm and the energy consumption cost objective function, in order to determine the high-quality equipment operating condition particles.

[0053] Figure 5 This is a flowchart of the steps in this application embodiment to analyze high-quality equipment condition particles according to the GA algorithm to determine crossover variant equipment condition particles. Detailed Implementation

[0054] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 5 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0055] Reference Figure 1 This application discloses a method for optimizing the energy efficiency of a refrigeration room, comprising the following steps:

[0056] Step S100: Obtain the basic forecast load and user equipment parameters.

[0057] Among them, the basic predicted load refers to the load of the chiller room to meet the cooling demand, which is predicted based on historical load data and current meteorological parameters. The processing terminal calls historical load data and meteorological parameters, inputs historical load data into the basic prediction fusion model to generate historical regular load prediction values, and then analyzes meteorological parameters to obtain meteorological sensitivity correction terms. Finally, the historical regular load prediction values ​​and meteorological sensitivity correction terms are weighted and summed according to dynamic load weights to obtain the predicted load.

[0058] Historical load data refers to the hourly measured records of cooling / heating load over the past 90 days, covering load fluctuation characteristics in different seasons and time periods, such as weekday morning peaks, nighttime off-peaks, and weekend stable periods. It is obtained by the processing terminal from the energy management system.

[0059] Meteorological parameters refer to meteorological data for the target forecast period, including the predicted outdoor temperature, predicted relative humidity, and predicted wind speed for the forecast period, which can be obtained directly through meteorological websites.

[0060] The basic predictive fusion model refers to a model that predicts future loads based on historical loads. In this embodiment, a fusion model of Long Short-Term Memory (LSTM) network and XGBoost is used. The LSM network includes an input layer, hidden layers, and an output layer. The input layer converts the data into a fixed-length vector and concatenates scene labels (e.g., weekdays) with numerical features and the load sequence. The hidden layer uses a forget gate to determine which historical information needs to be discarded. For example, if the current load differs significantly from the load of the previous few hours, the forget gate reduces the weight of past loads. The input gate updates the current state; for example, when a sudden temperature increase causes a load rise, the input gate strengthens the current load features. The output gate determines the current output. The current load trend is combined with scenario changes (such as weekdays and weekends) to generate predicted values. The output layer outputs the hidden state of the last time step as a temporal feature vector. XGBoost is used to optimize the output of the Long Short-Term Memory network. XGBoost supplements the average load, maximum load, and load change rate of the past 24 hours, and concatenates the temporal feature vector with the supplemented scenario features to form the final input. The objective function is to minimize the mean squared error between the predicted value and the true value. XGBoost automatically assigns the importance of different features by splitting tree nodes. For example, weekdays have higher weights. The predicted values ​​of the Long Short-Term Memory network are non-linearly adjusted, such as increasing the load prediction value on weekdays, and finally outputting the predicted load value.

[0061] The meteorological sensitivity correction term refers to the impact of meteorological data on the load. It is calculated by the processing terminal based on the temperature difference between the predicted temperature and the human comfort temperature, and then multiplied by the temperature sensitivity coefficient. The human comfort temperature is the ambient temperature at which humans feel comfortable, for example, 26 degrees Celsius. The temperature sensitivity coefficient refers to the load's response to temperature changes, obtained by fitting historical data according to specific application scenarios. Regression analysis can be used to determine the load variation patterns under different scenarios and temperatures; for example, the temperature sensitivity coefficient is higher in southern cities. The humidity sensitivity correction term is calculated by the processing terminal based on the humidity difference between the predicted relative humidity and the human comfort relative humidity, and then multiplied by the humidity sensitivity coefficient. The human comfort relative humidity is the humidity at which humans feel comfortable, for example, 50%. The humidity sensitivity coefficient refers to the intensity of the load's response to changes in humidity. It is obtained by fitting historical data according to specific application scenarios. Regression analysis can be used to determine the load variation patterns under different scenarios and humidity levels. For example, the humidity sensitivity coefficient is smaller in northern cities. The wind speed sensitivity correction term is obtained by multiplying the predicted wind speed and the wind speed sensitivity coefficient by the processing terminal. The wind speed sensitivity coefficient refers to the intensity of the load's response to changes in wind speed. It is obtained by fitting historical data according to specific application scenarios. Regression analysis can be used to determine the load variation patterns under different scenarios and wind speeds. For example, the wind speed sensitivity coefficient is larger in coastal cities. Finally, the sum of the temperature sensitivity correction term, humidity sensitivity correction term, and wind speed sensitivity correction term is calculated by the processing terminal, and then the sum is multiplied by the historical load prediction value to obtain the meteorological sensitivity correction term.

[0062] Dynamic load weighting refers to the weight of different dimensions of the predicted value in the total predicted value when predicting load. The specific value is determined by the operator. When the baseline value is used, the weights of historical regular load prediction values ​​and meteorological sensitive correction items decrease in sequence, and are dynamically adjusted by the processing terminal according to the scenario. For example, on weekdays, the weight of historical regular load prediction values ​​is increased, while the weight of meteorological sensitive correction items is decreased. In extreme weather, the weight of meteorological sensitive correction items is increased.

[0063] User equipment parameters refer to the intensity of behavior of different user groups on equipment within a building, including user group serial number, corresponding behavior type serial number, and corresponding behavior intensity variable. User group serial number includes employee serial number and visitor serial number. Adding user group serial number is mainly to consider that different user groups within the same building have different impact coefficients on load due to their different behavior patterns. For example, employees usually use energy-saving mode when using computers, while visitors may not have energy-saving mode when temporarily using equipment, resulting in different load increments for the same equipment power. Behavior type serial number includes office equipment usage serial number, socket equipment usage serial number, lighting control serial number, and window opening behavior serial number. Different behaviors have different impacts on load. Behavior intensity variable refers to the power density per unit area for different behaviors, usually monitored by sensors such as smart meters, human infrared sensors, and thermostats, statistical records such as attendance data and equipment operation logs, and indirectly estimated, such as estimating the percentage of power used based on lighting switch status.

[0064] Step S101: Analyze the user equipment parameters to determine the revised forecast load.

[0065] Among them, the corrected predicted load refers to the impact of different users' different behaviors within the building on the load, which is obtained by the processing terminal after analyzing user equipment parameters. Specific methods are described in [reference needed]. Figure 2 The steps.

[0066] Step S102: Analyze the basic forecast load and the revised forecast load to determine the final forecast load.

[0067] The final predicted load refers to the load that the chiller room will meet the cooling demand in the future period, which is obtained by summing the basic predicted load and the revised predicted load by the processing terminal.

[0068] Step S103: Construct a two-dimensional energy efficiency matrix of equipment operating conditions based on the final predicted load.

[0069] The equipment operating condition combination two-dimensional energy efficiency matrix refers to the combination of equipment and operating condition parameters that meet the final predicted load. The processing terminal constructs a decision variable space based on the total number of main units, cooling towers, water pumps and air conditioning terminals in the chiller room, enumerates all feasible candidate sets of equipment combinations + operating condition parameters, and uses the final predicted load as the boundary condition for optimization. It traverses each candidate combination and calculates whether it can meet the final predicted load. If it can meet the final predicted load, the candidate combination is included as part of the equipment operating condition combination two-dimensional energy efficiency matrix. Finally, all candidate sets are found to form the equipment operating condition combination two-dimensional energy efficiency matrix.

[0070] Step S104: Optimize the two-dimensional energy efficiency matrix of equipment operating condition combination according to the preset PSO-GA fusion algorithm to determine the recommended operating combination; the PSO-GA fusion algorithm includes the PSO algorithm and the GA algorithm.

[0071] The recommended operating combination refers to a combination of equipment and operating conditions that can meet the predicted load demand and has low energy consumption and cost. This combination is obtained by the processing terminal after optimizing the two-dimensional energy efficiency matrix of the equipment and operating conditions using the PSO-GA fusion algorithm. For specific methods, please refer to [link / reference]. Figure 3 The steps.

[0072] The PSO-GA fusion algorithm includes the PSO algorithm and the GA algorithm. The PSO algorithm is a particle swarm optimization algorithm used to find the equipment operating condition combinations with lower energy consumption and cost in the two-dimensional energy efficiency matrix of equipment operating condition combinations. The GA algorithm is a genetic algorithm used to cross-mutate the equipment operating condition combinations determined by the PSO algorithm to obtain new equipment operating condition combinations, and then continue to iterate.

[0073] Step S105: Adjust the operating status of the chiller room according to the recommended operating combination.

[0074] Once the recommended operating combination is determined, the processing terminal controls the equipment in the refrigeration room to operate according to the starting equipment corresponding to the recommended operating combination, as well as the specific load rate and speed of the equipment, thereby meeting the cooling demand while ensuring low energy consumption and cost.

[0075] Reference Figure 2 The steps for analyzing user equipment parameters to determine revised forecast loads include:

[0076] Step S200: Calculate the user equipment parameters according to the preset independent behavior algorithm to determine the independent behavior correction load; the user equipment parameters include the user group number, the corresponding behavior type number, and the corresponding behavior intensity variable.

[0077] Among them, the independent behavior-corrected load refers to the impact of the behavior of different user groups on the load from a single-dimensional perspective. It is obtained by the processing terminal after calculating the user equipment parameters according to the independent behavior algorithm. The expression of the independent behavior algorithm is as follows:

[0078] .

[0079] In the formula, Adjust the load for independent behavior. The total number of user group categories. For user group serial numbers, The total number of user behavior types For behavior type sequence number, The linear influence coefficient is... For behavioral intensity variables, This is a nonlinear influence coefficient. For the maximum lag time, For the time lag, This is the hysteresis attenuation coefficient.

[0080] The total number of user group categories refers to the total number of user groups that exist. In this embodiment of the application, 2 is used as an example, which means that there are two types of users: employees and visitors.

[0081] The total number of user behavior types refers to the total number of behaviors that exist. In this embodiment of the application, 4 is used as an example, which includes office equipment use, socket equipment use, lighting control and window opening behavior.

[0082] Maximum lag time refers to the longest duration of the impact of a behavior, which is determined based on the characteristics of the device or behavior. For example, a short-term behavior would take 30 minutes, while a long-term behavior would take 120 minutes.

[0083] Lag time refers to the duration of the impact of different behaviors on the load, that is, the duration of the impact after the behavior occurs.

[0084] The hysteresis attenuation coefficient refers to the residual impact coefficient of different users' behaviors on the current load when they occur before the hysteresis time. It usually decreases as the hysteresis time increases. It is calibrated by correlation analysis between historical behavior and hysteresis load. For example, to calculate the regression coefficient between oven usage 1 hour ago and current air conditioning load, if 1 oven was used 1 hour ago, and the current air conditioning load still increases by 0.3KW due to residual heat, then the hysteresis attenuation coefficient is 0.3.

[0085] First, identify the different user groups with user equipment parameters. Then, identify the behavior types and corresponding behavior intensity variables of different user groups. Finally, match the linear influence coefficient, nonlinear influence coefficient, and hysteresis decay coefficient corresponding to the user group and behavior in the database.

[0086] The linear impact coefficient refers to the linear impact coefficient of different users' behavior on the load. It is expressed as the average increase in load for every 1 unit increase in behavior intensity. It is calibrated through regression analysis of historical data. For example, if a linear regression is performed on the use of office equipment and the increase in load, the slope is the linear impact coefficient of the use of office equipment. For example, if the actual load increases by 0.95 kW for every 1 kW increase in the total power of office equipment, then the linear impact coefficient is 0.95.

[0087] The nonlinear impact coefficient refers to the nonlinear impact coefficient of different users' behavior on the load. It represents the rate of change of load increment when the intensity of behavior changes. The higher the intensity, the greater or smaller the load impact per unit increment. It is calibrated after discovering the nonlinear relationship between behavior intensity and load in historical data through polynomial regression fitting. For example, if the total power of employees' office equipment is 10KW and the nonlinear load increases by 2KW, then the nonlinear impact coefficient is 0.02.

[0088] Then, using an independent behavior algorithm, the product of the linear impact coefficient and the behavior type intensity variable is calculated to satisfy the requirement that the impact of a single behavior is linear. For example, if multiple behaviors occur in multiple spaces, the loads of the multiple behaviors can be summed. Next, the product of the nonlinear impact coefficient and the square of the behavior type intensity variable is calculated to satisfy the nonlinear requirement that multiple behaviors occur simultaneously. For example, when multiple behaviors occur simultaneously in the same space, the average load per person decreases. Thus, the linear and nonlinear impacts of the behaviors of different user groups on the load are summed to obtain the real-time impact value.

[0089] Finally, considering that the impact of user behavior on the load is not instantaneous but has a time lag or continuous decay, the lag attenuation coefficient corresponding to different lag times is calculated and multiplied by the corresponding behavior type intensity variable to obtain the lag impact value. The sum of the lag impact value and the real-time impact value is then calculated to obtain the independent behavior-corrected load.

[0090] Step S201: Calculate the user device parameters according to the preset interaction behavior algorithm to determine the interaction behavior correction load.

[0091] Among them, the interaction behavior correction load refers to the combined impact of user group behaviors on the load when they occur simultaneously. When different behaviors occur simultaneously, their combined impact is not equal to the sum of their individual impacts and needs to be corrected. For example, when one person uses office equipment and simultaneously uses the monitored socket equipment, the combined impact of the two behaviors is less than the sum of their individual impacts. Therefore, the interaction behavior algorithm is calculated based on the user equipment parameters. The expression of the interaction behavior algorithm is as follows:

[0092] ,

[0093] In the formula, Adjust the load for interactive behavior. For another behavior type sequence number, This represents the coefficient of influence of behavioral interactions.

[0094] First, identify the related behavior types in the user equipment parameters, such as using electrical outlets and using office equipment. Then, determine the individual intensity variables of these behaviors in the user equipment parameters and match the corresponding interaction influence coefficients in the database. The interaction influence coefficient refers to the influence coefficient on the load when two behaviors occur simultaneously for a user. When the joint influence value is greater than the sum of the individual influence values, the interaction influence coefficient is greater than 0; when the joint influence value is less than the sum of the individual influence values, the interaction influence coefficient is less than 0. Calibration is achieved through bivariate regression analysis. For example, the product of the number of electrical outlets used and the power of office equipment used is used as the independent variable, and its relationship with the load increment is fitted. For instance, if an employee uses 1 electrical outlet and uses 1 kW of office equipment, the interaction load increment is 0.25 kW, and the fitted interaction influence coefficient is 0.25. Then, the product of the behavior intensity variables of the two behaviors and the interaction influence coefficient is calculated to obtain the interaction behavior-corrected load. This prevents the load forecast from being underestimated or overestimated when two behaviors have an interaction effect and are calculated separately.

[0095] Step S202: Analyze the independent behavior-corrected load and the interactive behavior-corrected load to determine the corrected forecast load.

[0096] In this step, the corrected predicted load is the same as the corrected predicted load in step S101, and is obtained by the processing terminal by calculating the sum of the corrected load for independent behavior and the corrected load for interactive behavior.

[0097] Reference Figure 3 The steps for optimizing the two-dimensional energy efficiency matrix of equipment operating condition combinations based on the preset PSO-GA fusion algorithm to determine the recommended operating combinations include:

[0098] Step S300: Obtain the energy consumption cost objective function.

[0099] Among them, the energy consumption cost objective function refers to a function with energy consumption and cost as optimization objectives. The processing terminal substitutes the known quantities of the chiller room into the energy consumption calculation formula to determine the energy consumption function, and then calculates the product of the energy consumption function and the electricity price to obtain the cost function. Finally, the energy consumption function and the cost function are weighted and summed to obtain the energy consumption cost objective function.

[0100] Step S301: Calculate the two-dimensional energy efficiency matrix of equipment operating condition combination based on the PSO algorithm and energy consumption cost objective function to determine the high-quality equipment operating condition particles.

[0101] Among them, the high-quality equipment condition particles refer to the equipment condition combinations with lower energy consumption and cost in the two-dimensional energy efficiency matrix of equipment condition combinations. These are obtained by the processing terminal through optimization calculation of the two-dimensional energy efficiency matrix of equipment condition combinations based on the PSO algorithm and the energy consumption cost objective function. For specific methods, please refer to [reference needed]. Figure 4 The steps.

[0102] Step S302: Analyze the high-quality equipment condition particles according to the GA algorithm to determine the cross-mutated equipment condition particles.

[0103] Among them, the cross-mutated equipment condition particles refer to a subset of high-quality equipment condition particles after adjusting the equipment and condition parameters. These particles are obtained by the processing terminal after adjusting the high-quality equipment condition particles according to the GA algorithm. For specific methods, please refer to [link to relevant documentation]. Figure 5 The steps.

[0104] Step S303: Determine the optimal equipment condition particles based on the high-quality equipment condition particles and the cross-variant equipment condition particles.

[0105] Among them, the optimal equipment condition particle refers to the combination of equipment conditions to be optimized in the next round of iteration. The processing terminal defines the high-quality equipment condition particles and the cross-mutation equipment condition particles as the optimal equipment condition particles.

[0106] Step S304: Optimize the equipment condition particles according to the PSO-GA fusion algorithm to determine the recommended operating combination.

[0107] After determining the optimal equipment operating condition particles, the processing terminal continues to iterate the optimization of the optimal equipment operating condition particles according to the PSO-GA fusion algorithm until the final combination of equipment operating conditions is determined, which is the recommended operating combination.

[0108] Reference Figure 4 The steps for calculating the two-dimensional energy efficiency matrix of equipment operating condition combinations based on the PSO algorithm and energy consumption cost objective function to determine the high-quality equipment operating condition particles include:

[0109] Step S400: Calculate the two-dimensional energy efficiency matrix of equipment operating condition combination based on the energy consumption cost objective function to determine the particle fitness of equipment operating conditions.

[0110] Among them, the fitness of equipment operating conditions refers to the fitness of different combinations of equipment operating conditions. The higher the fitness, the lower the energy consumption and cost of the combination of equipment operating conditions. The processing terminal traverses the two-dimensional energy efficiency matrix of equipment operating conditions and substitutes the parameters of different combinations of equipment operating conditions into the energy consumption cost objective function to calculate the energy consumption cost value. The reciprocal of the energy consumption cost value is then calculated as the fitness of equipment operating conditions.

[0111] Step S401: Analyze the two-dimensional energy efficiency matrix of the equipment operating condition combination and the corresponding fitness of the equipment operating condition particles to determine the basic equipment operating condition particles.

[0112] Among them, the basic equipment condition particles refer to the equipment condition combinations with lower energy consumption and cost in the two-dimensional energy efficiency matrix of equipment condition combinations. The processing terminal selects the top 30 equipment condition combinations with the highest fitness from the two-dimensional energy efficiency matrix of equipment condition combinations to obtain the basic equipment condition particles.

[0113] Step S402: Determine whether the working condition particles of the basic equipment meet the requirements of the preset constraint conditions.

[0114] Among them, constraints refer to the conditions used to screen equipment operating condition combinations, such as flow rate, equipment speed, inlet and outlet temperature, and start-stop frequency must not exceed the set threshold. The requirement of constraints means that the constraints must be met.

[0115] The processing terminal determines whether the basic equipment operating condition particles meet the constraints, thereby determining whether the basic equipment operating condition particles are available.

[0116] Step S4021: If it does not meet the requirements, remove the basic equipment condition particles.

[0117] If the processing terminal determines that the basic equipment operating condition particle does not meet the constraint conditions, it indicates that the basic equipment operating condition particle is unavailable, and therefore the basic equipment operating condition particle is removed.

[0118] Step S4022: If the conditions are met, analyze the basic equipment operating condition particles to determine the high-quality equipment operating condition particles.

[0119] If the processing terminal determines that the basic equipment condition particles meet the constraints, it indicates that the basic equipment condition particles are available. Therefore, all basic equipment condition particles are sorted into high-quality equipment condition particles.

[0120] Reference Figure 5 The steps for analyzing high-quality equipment condition particles using the GA algorithm to determine crossover variant equipment condition particles include:

[0121] Step S500: Analyze the high-quality equipment condition particles to determine the selected equipment condition particles.

[0122] Among them, the selected equipment condition particles refer to the particles with the top 20 fitness among the high-quality equipment condition particles, which are selected by the processing terminal from the high-quality equipment condition particles.

[0123] Step S501: Cross the selected equipment condition particles to generate cross equipment condition particles.

[0124] Among them, the cross-equipment condition particle refers to the equipment condition combination obtained by fusing high-quality parameters from different equipment condition combinations. The processing terminal randomly selects two particles from the selected equipment condition particles and generates the cross-equipment condition particle by exchanging some parameter fragments. Single-point crossover can be used. First, the crossover probability is determined, and the optimal parameters in the examples are retained. Then, a crossover point is randomly selected. One crossover point is randomly selected from the gaps in the equipment condition parameters. The parameter fragments on the right side of the crossover point of the two examples are exchanged to obtain two new particles. Then, it is checked whether the new particles meet the constraints and corrected to obtain the cross-equipment condition particle.

[0125] Step S502: Mutate the cross-equipment condition particles to generate cross-mutated equipment condition particles.

[0126] In this step, the cross-mutated equipment condition particles are the same as those in step S302. The processing terminal makes small-amplitude random adjustments to one or more parameters in the cross-mutated equipment condition particles to break the local optimal parameter combination. Site mutation can be used. First, the mutation probability is determined and high-quality parameters are retained. Then, one parameter is randomly selected from the particles as the mutation site. The mutation site is fine-tuned with a fixed amplitude, and the constraints are checked and corrected to obtain the cross-mutated equipment condition particles.

[0127] Based on the same inventive concept, embodiments of this application provide a chiller room energy efficiency optimization system, including:

[0128] The acquisition module is used to acquire the basic forecast load, user equipment parameters, and energy consumption cost objective function;

[0129] A memory for storing a program for optimizing the energy efficiency of a refrigeration room;

[0130] The processor and memory can load and execute programs to implement a method for optimizing the energy efficiency of a cooling room.

[0131] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0132] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed to provide a method for optimizing energy efficiency in a cooling room.

[0133] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0134] Based on the same inventive concept, embodiments of this application provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to optimize the energy efficiency of a refrigeration room.

[0135] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0136] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for optimizing energy efficiency in a refrigeration room, characterized in that, include: Obtain basic forecast load and user equipment parameters; Analyze user equipment parameters to determine the revised forecast load; The basic forecast load and the revised forecast load are analyzed to determine the final forecast load; A two-dimensional energy efficiency matrix of equipment operating conditions is constructed based on the final predicted load. The two-dimensional energy efficiency matrix of equipment operating condition combinations is optimized according to the preset PSO-GA fusion algorithm to determine the recommended operating combination; the PSO-GA fusion algorithm includes the PSO algorithm and the GA algorithm. The operating status of the chiller room is adjusted according to the recommended operating combination; The steps for analyzing user equipment parameters to determine revised forecast loads include: The user equipment parameters are calculated based on a preset independent behavior algorithm to determine the independent behavior correction load; the user equipment parameters include the user group number, the corresponding behavior type number, and the corresponding behavior intensity variable; The user device parameters are calculated based on a preset interaction behavior algorithm to determine the interaction behavior correction load; The modified loads for independent behavior and the modified loads for interactive behavior are analyzed to determine the modified forecast loads; The expression for the independent behavior algorithm is: ， In the formula, Adjust the load for independent behavior. The total number of user group categories. User group number The total number of user behavior types. For behavior type sequence number, The linear influence coefficient is... For behavioral intensity variables, This is a nonlinear influence coefficient. For the maximum lag time, For the time lag, This is the hysteresis attenuation coefficient.

2. The energy efficiency optimization method for a refrigeration room according to claim 1, characterized in that, The expression for the interaction behavior algorithm is: ， In the formula, Adjust the load for interactive behavior. For another behavior type sequence number, This represents the coefficient of influence of behavioral interactions.

3. The energy efficiency optimization method for a refrigeration room according to claim 1, characterized in that, The steps for optimizing the two-dimensional energy efficiency matrix of equipment operating condition combinations based on the preset PSO-GA fusion algorithm to determine the recommended operating combinations include: Obtain the energy consumption cost objective function; The two-dimensional energy efficiency matrix of equipment operating condition combinations is calculated based on the PSO algorithm and the energy consumption cost objective function to determine the high-quality equipment operating condition particles. The high-quality equipment condition particles are analyzed using the GA algorithm to determine the cross-variant equipment condition particles. The optimal equipment condition particles are determined based on the high-quality equipment condition particles and the cross-variant equipment condition particles. The PSO-GA fusion algorithm is used to further optimize the equipment operating condition particles to determine the recommended operating combination.

4. The energy efficiency optimization method for a refrigeration room according to claim 3, characterized in that, The steps for calculating the two-dimensional energy efficiency matrix of equipment operating condition combinations based on the PSO algorithm and energy consumption cost objective function to determine the high-quality equipment operating condition particles include: The two-dimensional energy efficiency matrix of equipment operating condition combination is calculated based on the energy consumption cost objective function to determine the particle fitness of equipment operating conditions. The two-dimensional energy efficiency matrix of equipment operating condition combinations and the corresponding fitness of equipment operating condition particles are analyzed to determine the basic equipment operating condition particles. Determine whether the working conditions of the basic equipment particles meet the requirements of the preset constraints; If it does not meet the requirements, the basic equipment condition particles will be removed. If the conditions are met, the basic equipment operating condition particles are analyzed to determine the high-quality equipment operating condition particles.

5. The energy efficiency optimization method for a refrigeration room according to claim 3, characterized in that, The steps for analyzing high-quality equipment condition particles using the GA algorithm to determine crossover variant equipment condition particles include: Analyze high-quality equipment operating condition particles to determine the selection of equipment operating condition particles; Cross the selected equipment condition particles to generate cross equipment condition particles; Mutate the cross-equipment condition particles to generate cross-mutated equipment condition particles.

6. A chiller room energy efficiency optimization system, characterized in that, include: The acquisition module is used to acquire basic forecast load and user equipment parameters; A memory for storing a program for a method of optimizing energy efficiency in a refrigeration room as described in any one of claims 1 to 5; The processor and the program in the memory can be loaded and executed by the processor to implement the energy efficiency optimization method for a refrigeration room as described in any one of claims 1 to 5.

7. A smart terminal, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 5, which is a method for optimizing the energy efficiency of a refrigeration room.

Images