Energy consumption optimization method, system, device and medium based on industrial internet of things

Abstract

The application discloses an energy consumption optimization method, system, device and medium based on an industrial Internet of Things, relates to the technical field of the industrial Internet of Things, and the method comprises the following steps: acquiring historical setting parameter data, historical energy consumption data and energy consumption rating data of each processing device in a production line; constructing an energy consumption prediction model according to the historical setting parameter data and the historical energy consumption data, and constructing an energy consumption grade prediction model according to the historical energy consumption data and the energy consumption rating data; acquiring a target energy consumption grade, setting an energy consumption limitation condition, calling the energy consumption grade prediction model, and searching according to a gradient descent algorithm to obtain a target energy consumption parameter; acquiring a target setting parameter of a target processing device, setting a setting parameter constraint condition, calling the energy consumption prediction model, and searching based on a particle swarm optimization algorithm to determine optimal setting parameters of non-target processing devices. The application has the effect of improving the energy consumption utilization efficiency of the production line.

Description

Technical Field

[0001] This application relates to the technical field of industrial Internet of Things (IIoT), and in particular to energy consumption optimization methods, systems, devices, and media based on IIoT. Background Technology

[0002] Currently, energy consumption optimization in production lines mainly relies on traditional monitoring methods and empirical rules. This involves manually adjusting equipment operating parameters by observing real-time energy consumption data in order to reduce energy waste. Some companies have introduced automatic control systems that use sensors to collect data and then automatically adjust equipment according to preset rules. However, these adjustments are often static and struggle to cope with constantly changing operating conditions during production, thus having certain limitations. Furthermore, many companies still rely on manual monitoring and intervention, lacking systematic and intelligent scheduling methods, resulting in difficulties in significantly improving energy consumption levels.

[0003] Although some technologies have begun to incorporate big data and intelligent control, they still face problems such as insufficient data, limited optimization methods, and difficulty in coordinating equipment in practical applications. This limits the overall energy efficiency improvement potential of production lines, making it difficult to achieve refined and dynamic energy consumption management. Consequently, the energy utilization efficiency of production lines remains low, failing to fully utilize the potential of equipment and resulting in a low level of production intelligence, which is insufficient to meet the automation and intelligence requirements of modern industry for energy conservation. Summary of the Invention

[0004] To improve the energy efficiency of production lines and thus enhance their intelligence level, this application provides an energy consumption optimization method, system, equipment, and medium based on the Industrial Internet of Things.

[0005] Firstly, this application provides an energy consumption optimization method based on the Industrial Internet of Things, employing the following technical solution:

[0006] An energy consumption optimization method based on the Industrial Internet of Things (IIoT) is applied to an IIoT system, which includes a management platform, a sensor network platform, and an object platform connected in sequence. The method is executed by the management platform and includes:

[0007] Acquire historical setting parameter data, historical energy consumption data, and energy consumption rating data for each processing equipment in the production line;

[0008] An energy consumption prediction model for the production line is constructed based on the historical set parameter data and the historical energy consumption data, and an energy consumption level prediction model for the production line is constructed based on the historical energy consumption data and the energy consumption rating data.

[0009] Obtain the target energy consumption level, set energy consumption limits, call the energy consumption level prediction model, and search according to the gradient descent algorithm to obtain the target energy consumption parameters, wherein the energy consumption prediction level of the target energy consumption parameters is consistent with the target energy consumption level;

[0010] Obtain the target setting parameters of the target processing equipment, set the setting parameter constraints, call the energy consumption prediction model, and search based on the particle swarm optimization algorithm to determine the optimal setting parameters of the non-target processing equipment. Among them, the energy consumption prediction parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters.

[0011] By adopting the above technical solution, historical setting parameter data, historical energy consumption data, and energy consumption rating data of each processing equipment in the production line are first obtained. Then, an energy consumption prediction model for the production line is constructed based on the historical setting parameter data and historical energy consumption data, and an energy consumption level prediction model is constructed based on the historical energy consumption data and energy consumption rating data. Next, the target energy consumption level of the production line is obtained, energy consumption constraints are set, the energy consumption level prediction model is called, and the target energy consumption parameters are obtained by searching using the gradient descent algorithm. The predicted energy consumption level of the target energy consumption parameters is consistent with the target energy consumption level. Then, the target setting parameters of the target processing equipment are obtained, setting constraints are set, the energy consumption prediction model is called, and the optimal setting parameters of non-target processing equipment are determined by searching using the particle swarm optimization algorithm. The predicted energy consumption parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters. Through the above method, refined, dynamic, and intelligent management of energy consumption is achieved, which is conducive to improving energy utilization efficiency, reducing energy waste, improving the intelligence level of the production line, and realizing multi-equipment coordinated optimization, improving the overall energy utilization balance and stability, and promoting energy conservation, consumption reduction, and green production of the production line.

[0012] Optionally, the step of constructing the energy consumption prediction model of the production line based on the historical setting parameter data and the historical energy consumption data includes:

[0013] The historical setting parameter data and the historical energy consumption data are used as the first model training data, and the first model training data is divided into a first training set and a first test set according to a first preset ratio.

[0014] The hyperparameters of the pre-built decision tree model are set according to the random network search algorithm, and the first root mean square error (RMSE) and the first coefficient of determination (R²) are set. 2 As the primary evaluation indicator;

[0015] The pre-built decision tree model is trained based on the first training set to obtain a trained decision tree model;

[0016] The trained decision tree model is tested according to the first test set, and the corresponding error is determined according to the first evaluation index. If it is within the first preset range, the trained decision tree model is used as the energy consumption prediction model of the production line.

[0017] By adopting the above technical solution, in order to construct an energy consumption prediction model for the production line, historical set parameter data and historical energy consumption data are used as the first model training data. The first model training data is divided into a first training set and a first test set according to a first preset ratio. Then, the hyperparameters of the pre-constructed decision tree model are set according to a random network search algorithm, and the first root mean square error (RMSE) and the first coefficient of determination (R²) are set. 2 The first evaluation metric is used as the basis for training the pre-built decision tree model based on the first training set to obtain the trained decision tree model. Finally, the trained decision tree model is tested based on the first test set, and the corresponding error is judged according to the first evaluation metric to determine whether it is within the first preset range. If it is within the first preset range, the trained decision tree model is used as the energy consumption prediction model for the production line.

[0018] Optionally, the step of constructing an energy consumption level prediction model for the production line based on the historical energy consumption data and the energy consumption rating data includes:

[0019] The historical energy consumption data and the energy consumption rating data are used as training data for the second model, and the training data for the second model is divided into a second training set and a second test set according to a second preset ratio.

[0020] The hyperparameters of the pre-built neural network model are set according to the random network search algorithm, and the second root mean square error (RMSE) and the second coefficient of determination (R²) are set. 2 As a second evaluation indicator;

[0021] The pre-built neural network model is trained according to the second training set to obtain a trained neural network model;

[0022] The trained neural network model is tested according to the second test set, and the corresponding error is determined according to the second evaluation index to be within the second preset range. If it is, the trained neural network model is used as the energy consumption level prediction model of the production line.

[0023] By adopting the above technical solution, in order to construct a production line energy consumption level prediction model, historical energy consumption data and energy consumption rating data are used as training data for the second model. The training data is divided into a second training set and a second test set according to a second preset ratio. Then, the hyperparameters of the pre-constructed neural network model are set according to a random network search algorithm, and the second root mean square error (RMSE) and the second coefficient of determination (R²) are set. 2 As the second evaluation metric, the pre-built neural network model is then trained based on the second training set to obtain a trained neural network model. Finally, the trained neural network model is tested based on the second test set, and the corresponding error is judged according to the second evaluation metric to determine whether it is within the second preset range. If it is within the second preset range, the trained neural network model is used as the energy consumption level prediction model for the production line.

[0024] Optionally, the step of calling the energy consumption level prediction model and searching according to the gradient descent algorithm to obtain the target energy consumption parameters includes:

[0025] A first search space is set according to the energy consumption limit conditions, a first initial value is set according to the first search space, and a first initial search direction is set.

[0026] Based on the iteration point and search direction The iteration is performed within the first search space, wherein... For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient at the (k+1)th iteration point;

[0027] During the iteration process, for each iteration point, energy consumption level is predicted according to the energy consumption level prediction model, and it is determined whether the corresponding predicted energy consumption level is consistent with the target energy consumption level. If so, the corresponding iteration point is used as the target energy consumption parameter.

[0028] By adopting the above technical solution, in order to obtain the target energy consumption parameters, a first search space is set according to the energy consumption constraints, a first initial value is set according to the first search space, and a first initial search direction is set. Then, based on the iteration points... and search direction Iterate within the first search space, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. Let be the gradient at the (k+1)th iteration point. During the iteration process, for each iteration point, energy consumption level is predicted according to the energy consumption level prediction model, and it is determined whether the corresponding predicted energy consumption level is consistent with the target energy consumption level. If they are consistent, the corresponding iteration point is taken as the target energy consumption parameter.

[0029] Optionally, the step of calling the energy consumption prediction model and searching within the set parameter constraints based on the particle swarm optimization algorithm to determine the optimal set parameters for the non-target processing equipment includes:

[0030] A second search space is set according to the set parameter constraints, and a set of particles is set according to the second search space and the target set parameters, and the velocity of each particle is set.

[0031] Update formula based on location and speed update formula

[0032] The process iterates over each particle within the second search space, where... For the (k+1)th iteration point, For the k-th iteration point, The velocity in the k-th iteration, Let w be the velocity in the (k+1)th iteration, and w be the inertia weight. For cognitive factors, A random number between 0 and 1 To be the best location found during the search process, As a social learning factor, A random number between 0 and 1 The best position found globally. For random number functions, The amplitude of the disturbance;

[0033] During the iteration process, for each iteration of a group of particles, the energy consumption of each particle's iteration point is predicted according to the energy consumption prediction model, and it is determined whether the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than a preset value. If so, the optimal setting parameters of the non-target processing equipment are determined according to the corresponding iteration point, and the iteration process ends.

[0034] Otherwise, n parameter points are randomly selected in the second search space according to the target setting parameters, and the distance between the energy consumption prediction parameter corresponding to each parameter point and the target energy consumption parameter is calculated.

[0035] The shortest distance is determined based on the distance, and the optimal setting parameters for the non-target processing equipment are determined based on the parameter point corresponding to the shortest distance.

[0036] By adopting the above technical solution, in order to determine the optimal setting parameters of the non-target processing equipment, a second search space is set according to the setting parameter constraints, and a set of particles is set according to the second search space and the target setting parameters, and the velocity of each particle is set. Then, the position update formula is used. and speed update formula The process is iterated for each particle separately within the second search space, where... For the (k+1)th iteration point, For the k-th iteration point, The velocity in the k-th iteration, Let w be the velocity in the (k+1)th iteration, and w be the inertia weight. For cognitive factors, A random number between 0 and 1 To be the best location found during the search process, As a social learning factor, A random number between 0 and 1 The best position found globally. For random number functions, The disturbance amplitude is defined as follows: During the iteration process, for each iteration of a group of particles, energy consumption is predicted for each particle's iteration point according to the energy consumption prediction model. It is then determined whether the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than a preset value. If the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than the preset value, the optimal setting parameters for the non-target processing equipment are determined based on the corresponding iteration point, and the iteration process ends. Otherwise, n parameter points are randomly selected in the second search space based on the target setting parameters, and the distance between the energy consumption prediction parameter and the target energy consumption parameter corresponding to each parameter point is calculated. The shortest distance is then determined based on the distance, and the optimal setting parameters for the non-target processing equipment are determined based on the parameter point corresponding to the shortest distance.

[0037] Optionally, the method further includes:

[0038] Obtain the current setting parameters of the production line, wherein the current setting parameters include at least one or more of the following for each processing equipment: rotational speed, feed rate, cutting speed, vibration frequency, stress characteristics, pulse parameters, coolant flow rate, and coolant pressure;

[0039] The current set parameters are input into the energy consumption prediction model to obtain the predicted value of the current energy consumption parameters corresponding to the production line.

[0040] By adopting the above technical solution, in order to predict energy consumption based on the energy consumption prediction model, the current setting parameters of the production line are obtained. The current setting parameters include at least one or more of the following: rotational speed, feed rate, cutting speed, vibration frequency, stress characteristics, pulse parameters, coolant flow rate, and coolant pressure of each processing equipment. Then, the current setting parameters are input into the energy consumption prediction model to obtain the predicted value of the current energy consumption parameters of the production line.

[0041] Optionally, the method further includes:

[0042] The current energy consumption parameters of the production line are obtained and input into the energy consumption level prediction model to obtain the current energy consumption level corresponding to the production line.

[0043] By adopting the above technical solution, in order to make predictions based on the energy consumption level prediction model, the current energy consumption parameters of the production line are obtained, and the current energy consumption parameters are input into the energy consumption level prediction model to obtain the current energy consumption level corresponding to the production line.

[0044] Secondly, this application also provides an energy consumption optimization system based on the Industrial Internet of Things, which adopts the following technical solution:

[0045] An energy consumption optimization system based on the Industrial Internet of Things (IIoT) includes a management platform, a sensor network platform, and an object platform that are sequentially connected in communication. The management platform is configured with:

[0046] The data acquisition module is used to acquire historical setting parameter data, historical energy consumption data, and energy consumption rating data corresponding to the historical energy consumption data of each processing equipment in the production line. The historical energy consumption data includes at least two of the following: historical total energy consumption data, historical effective energy consumption data, and historical loss energy consumption data.

[0047] The model building module is used to build an energy consumption prediction model for the production line based on the historical set parameter data and the historical energy consumption data, and to build an energy consumption level prediction model for the production line based on the historical energy consumption data and the energy consumption rating data.

[0048] The energy consumption parameter determination module is used to obtain the target energy consumption level of the production line, set the energy consumption limit conditions of the production line, call the energy consumption level prediction model, and search within the energy consumption limit conditions according to the gradient descent algorithm to obtain the target energy consumption parameter, wherein the energy consumption prediction level of the target energy consumption parameter is consistent with the target energy consumption level.

[0049] The parameter setting module is used to obtain the target setting parameters of the target processing equipment, set the setting parameter constraints, call the energy consumption prediction model, and search within the setting parameter constraints based on the particle swarm optimization algorithm to determine the optimal setting parameters of the non-target processing equipment. The processing equipment includes the target processing equipment and the non-target processing equipment, and the optimal setting parameters and the energy consumption prediction parameters corresponding to the target setting parameters are closest to the target energy consumption parameters.

[0050] Thirdly, this application also provides a computer device, which adopts the following technical solution:

[0051] A computer device includes a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the method described in the first aspect.

[0052] Fourthly, this application also provides a computer-readable storage medium, which adopts the following technical solution:

[0053] A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the method described in the first aspect.

[0054] In summary, this application includes at least the following beneficial technical effects: First, it acquires historical setting parameter data, historical energy consumption data, and energy consumption rating data for each processing equipment in the production line. Then, it constructs an energy consumption prediction model for the production line based on the historical setting parameter data and historical energy consumption data, and constructs an energy consumption level prediction model for the production line based on the historical energy consumption data and energy consumption rating data. Next, it obtains the target energy consumption level of the production line, sets energy consumption constraints for the production line, calls the energy consumption level prediction model, and searches using the gradient descent algorithm to obtain the target energy consumption parameters. The predicted energy consumption level of the target energy consumption parameters is consistent with the target energy consumption level. Then, the target setting parameters of the target processing equipment are obtained, setting parameter constraints are set, the energy consumption prediction model is called, and the optimal setting parameters of non-target processing equipment are determined based on the particle swarm optimization algorithm. Among them, the energy consumption prediction parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters. Through the above method, refined, dynamic, and intelligent management of energy consumption is achieved, which is conducive to improving energy utilization efficiency, reducing energy waste, improving the intelligence level of the production line, and realizing multi-equipment coordinated optimization, improving the balance and stability of overall energy utilization, and promoting energy conservation, consumption reduction, and green production of the production line. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the overall process of an embodiment of this application.

[0056] Figure 2 This is a structural diagram of one application scenario of the system in this application embodiment.

[0057] Figure 3 This is a structural diagram of another application scenario of the system according to an embodiment of this application.

[0058] Figure 4 This is a structural block diagram of the computer device described in this application. Detailed Implementation

[0059] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figure 1-4 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.

[0060] This application discloses an energy consumption optimization method based on the Industrial Internet of Things.

[0061] Reference Figure 1 An energy consumption optimization method based on the Industrial Internet of Things (IIoT) is applied to an IIoT system. The IIoT system includes a management platform, a sensor network platform, and an object platform that are sequentially connected via communication. The method is executed by the management platform and includes:

[0062] Step S11: Obtain historical setting parameter data, historical energy consumption data, and energy consumption rating data for each processing equipment in the production line.

[0063] Historical energy consumption data includes at least two of the following: historical total energy consumption data, historical effective energy consumption data, and historical energy loss data.

[0064] It should be noted that a production line typically contains multiple processing equipment, and the setting parameters of a production line consist of the setting parameters of each processing equipment in that production line. The historical energy consumption data in this application refers to the historical energy consumption data of the entire production line. The energy consumption rating data in this application is the data obtained by developers or evaluators based on the historical energy consumption data, or the data obtained by rating the historical energy consumption data according to pre-set rating rules. In addition, the total energy consumption of the production line mainly includes two parts: effective energy consumption and loss energy consumption.

[0065] Step S12: Construct an energy consumption prediction model for the production line based on historical setting parameter data and historical energy consumption data, and construct an energy consumption level prediction model for the production line based on historical energy consumption data and energy consumption rating data.

[0066] Step S13: Obtain the target energy consumption level, set energy consumption constraints, call the energy consumption level prediction model, and search according to the gradient descent algorithm to obtain the target energy consumption parameters. The predicted energy consumption level of the target energy consumption parameters is consistent with the target energy consumption level.

[0067] It should be noted that the target energy consumption level is set by the staff based on the current production situation. For example, the energy consumption level is AE, which means that the power consumption is increasing. When the power supply is tight, the target energy consumption level can be set to A to save power and reduce costs. At this time, the processing speed of the production line may be reduced.

[0068] Step S14: Obtain the target setting parameters of the target processing equipment, set the setting parameter constraints, call the energy consumption prediction model, and search based on the particle swarm optimization algorithm to determine the optimal setting parameters of the non-target processing equipment. Among them, the energy consumption prediction parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters.

[0069] It should be noted that the target processing equipment is one or more processing equipment selected by the staff. These processing equipment require parameters to be set manually for certain reasons (such as production plan, working hours), i.e. target setting parameters. Each processing equipment in the production line consists of target processing equipment and non-target processing equipment.

[0070] In the above implementation, historical setting parameter data, historical energy consumption data, and energy consumption rating data of each processing equipment in the production line are first acquired. Then, an energy consumption prediction model for the production line is constructed based on the historical setting parameter data and historical energy consumption data, and an energy consumption level prediction model for the production line is constructed based on the historical energy consumption data and energy consumption rating data. Next, the target energy consumption level of the production line is obtained, energy consumption constraints are set, the energy consumption level prediction model is invoked, and a search is performed using the gradient descent algorithm to obtain the target energy consumption parameters. The predicted energy consumption level of the target energy consumption parameters is consistent with the target energy consumption level. Then, the target setting parameters of the target processing equipment are obtained, setting parameter constraints are set, the energy consumption prediction model is invoked, and a search is performed using the particle swarm optimization algorithm to determine the optimal setting parameters of non-target processing equipment. The predicted energy consumption parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters. Through the above method, refined, dynamic, and intelligent management of energy consumption is achieved, which is conducive to improving energy utilization efficiency, reducing energy waste, improving the intelligence level of the production line, and realizing multi-equipment coordinated optimization, improving the balance and stability of overall energy utilization, and promoting energy conservation, consumption reduction, and green production of the production line.

[0071] As a further implementation of the method, the step of constructing an energy consumption prediction model for the production line based on historical set parameter data and historical energy consumption data includes:

[0072] Step S21: Use historical setting parameter data and historical energy consumption data as the first model training data, and divide the first model training data into a first training set and a first test set according to a first preset ratio.

[0073] It should be noted that the first preset ratio can be 7:3 or 8:2.

[0074] Step S22: Set the hyperparameters of the pre-built decision tree model according to the random network search algorithm, and set the first root mean square error (RMSE) and the first coefficient of determination (R²). 2 It serves as the primary evaluation indicator.

[0075] It's important to note that Random Search is a method used for hyperparameter optimization. In machine learning models, hyperparameters are parameters that need to be manually set during model training, such as the learning rate, regularization coefficient, and tree depth. The main idea of ​​Random Search is to randomly select a set of parameter combinations from a given hyperparameter space, evaluate these combinations, and select the parameters with the best performance.

[0076] Step S23: Train the pre-built decision tree model based on the first training set to obtain the trained decision tree model.

[0077] Step S24: Test the trained decision tree model according to the first test set, and determine whether the corresponding error is within the first preset range according to the first evaluation index. If so, use the trained decision tree model as the energy consumption prediction model for the production line.

[0078] In the above embodiments, in order to construct an energy consumption prediction model for the production line, historical set parameter data and historical energy consumption data are used as the first model training data. The first model training data is divided into a first training set and a first test set according to a first preset ratio. Then, the hyperparameters of the pre-constructed decision tree model are set according to a random network search algorithm, and the first root mean square error (RMSE) and the first coefficient of determination (R²) are set. 2 The first evaluation metric is used as the basis for training the pre-built decision tree model based on the first training set to obtain the trained decision tree model. Finally, the trained decision tree model is tested based on the first test set, and the corresponding error is judged according to the first evaluation metric to determine whether it is within the first preset range. If it is within the first preset range, the trained decision tree model is used as the energy consumption prediction model for the production line.

[0079] As a further implementation of the method, the step of constructing an energy consumption level prediction model for the production line based on historical energy consumption data and energy consumption rating data includes:

[0080] Step S31: Use historical energy consumption data and energy consumption rating data as training data for the second model, and divide the training data of the second model into a second training set and a second test set according to a second preset ratio.

[0081] It should be noted that the second preset ratio can be 7:3 or 8:2.

[0082] Step S32: Set the hyperparameters of the pre-built neural network model according to the random network search algorithm, and set the second root mean square error (RMSE) and the second coefficient of determination (R²). 2 As a second evaluation indicator.

[0083] Step S33: Train the pre-built neural network model according to the second training set to obtain the trained neural network model.

[0084] Step S34: Test the trained neural network model according to the second test set, and determine whether the corresponding error is within the second preset range according to the second evaluation index. If so, use the trained neural network model as the energy consumption level prediction model for the production line.

[0085] In the above implementation, in order to construct an energy consumption level prediction model for the production line, historical energy consumption data and energy consumption rating data are used as training data for the second model. The training data is divided into a second training set and a second test set according to a second preset ratio. Then, the hyperparameters of the pre-constructed neural network model are set according to a random network search algorithm, and the second root mean square error (RMSE) and the second coefficient of determination (R²) are set. 2 As the second evaluation metric, the pre-built neural network model is then trained based on the second training set to obtain a trained neural network model. Finally, the trained neural network model is tested based on the second test set, and the corresponding error is judged according to the second evaluation metric to determine whether it is within the second preset range. If it is within the second preset range, the trained neural network model is used as the energy consumption level prediction model for the production line.

[0086] As a further implementation of the method, the step of calling the energy consumption level prediction model and searching according to the gradient descent algorithm to obtain the target energy consumption parameters includes:

[0087] Step S41: Set a first search space according to the energy consumption limit, set a first initial value according to the first search space, and set a first initial search direction.

[0088] Step S42, based on the iteration point and search direction Iterate within the first search space.

[0089] in, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. This is the gradient at the (k+1)th iteration point.

[0090] Step S43: During the iteration process, for each iteration point, energy consumption level is predicted according to the energy consumption level prediction model, and it is determined whether the corresponding predicted energy consumption level is consistent with the target energy consumption level. If so, the corresponding iteration point is used as the target energy consumption parameter.

[0091] In the above implementation, in order to obtain the target energy consumption parameters, a first search space is set according to the energy consumption constraints, a first initial value is set according to the first search space, and a first initial search direction is set. Then, based on the iteration points... and search direction Iterate within the first search space, where, For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. Let be the gradient at the (k+1)th iteration point. During the iteration process, for each iteration point, energy consumption level is predicted according to the energy consumption level prediction model, and it is determined whether the corresponding predicted energy consumption level is consistent with the target energy consumption level. If they are consistent, the corresponding iteration point is taken as the target energy consumption parameter.

[0092] As a further implementation of the method, the step of calling an energy consumption prediction model and searching based on a particle swarm optimization algorithm to determine the optimal setting parameters for non-target processing equipment includes:

[0093] Step S51: Set a second search space according to the set parameter constraints, and set a group of particles according to the second search space and the target setting parameters, and set the velocity of each particle.

[0094] Step S52, update the formula based on position and speed update formula

[0095] The process is repeated for each particle in the second search space.

[0096] in, For the (k+1)th iteration point, For the k-th iteration point, The velocity in the k-th iteration, Let w be the velocity in the (k+1)th iteration, and w be the inertia weight. For cognitive factors, A random number between 0 and 1 To be the best location found during the search process, As a social learning factor, A random number between 0 and 1 The best position found globally. For random number functions, The amplitude of the disturbance.

[0097] It should be noted that, Used to indicate that the particle maintains its inertia and continues to move in the current direction (exploration). Used to guide particles toward their historical best point (cognitive behavior). Used to guide particles to move toward the global optimum of the entire population (social behavior). The random perturbation is used to increase the diversity of the search and avoid getting trapped in local optima.

[0098] Step S53: During the iteration process, for each iteration of a group of particles, energy consumption is predicted for each iteration point of each particle according to the energy consumption prediction model, and it is determined whether the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than the preset value. If so, the optimal setting parameters of the non-target processing equipment are determined according to the corresponding iteration point, and the iteration process ends.

[0099] Step S54: Otherwise, randomly select n parameter points in the second search space according to the target setting parameters, and calculate the distance between the energy consumption prediction parameter and the target energy consumption parameter corresponding to each parameter point.

[0100] Step S55: Determine the shortest distance based on the distance, and determine the optimal setting parameters for the non-target processing equipment based on the parameter points corresponding to the shortest distance.

[0101] In the above embodiments, in order to determine the optimal setting parameters of the non-target processing equipment, a second search space is set according to the setting parameter constraints, and a set of particles is set according to the second search space and the target setting parameters, and the velocity of each particle is set. Then, the position update formula is used. and speed update formula

[0102] The process is iterated for each particle separately within the second search space, where... For the (k+1)th iteration point, For the k-th iteration point, The velocity in the k-th iteration, Let w be the velocity in the (k+1)th iteration, and w be the inertia weight. For cognitive factors, A random number between 0 and 1 To be the best location found during the search process, As a social learning factor, A random number between 0 and 1 The optimal position found globally. For random number functions, The disturbance amplitude is defined as follows: During the iteration process, for each iteration of a group of particles, energy consumption is predicted for each particle's iteration point according to the energy consumption prediction model. It is then determined whether the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than a preset value. If the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than the preset value, the optimal setting parameters for the non-target processing equipment are determined based on the corresponding iteration point, and the iteration process ends. Otherwise, n parameter points are randomly selected in the second search space based on the target setting parameters, and the distance between the energy consumption prediction parameter and the target energy consumption parameter corresponding to each parameter point is calculated. The shortest distance is then determined based on the distance, and the optimal setting parameters for the non-target processing equipment are determined based on the parameter point corresponding to the shortest distance.

[0103] As a further implementation of the method, the method also includes:

[0104] Step S61: Obtain the current setting parameters of the production line.

[0105] The current settings include at least one or more of the following parameters for each processing equipment: rotational speed, feed rate, cutting speed, vibration frequency, stress characteristics, pulse parameters, coolant flow rate, and coolant pressure.

[0106] Step S62: Input the current set parameters into the energy consumption prediction model to obtain the predicted value of the current energy consumption parameters corresponding to the production line.

[0107] In the above implementation, in order to predict energy consumption based on the energy consumption prediction model, the current setting parameters of the production line are obtained. The current setting parameters include at least one or more of the following: rotational speed, feed rate, cutting speed, vibration frequency, stress characteristics, pulse parameters, coolant flow rate, and coolant pressure of each processing equipment. Then, the current setting parameters are input into the energy consumption prediction model to obtain the predicted value of the current energy consumption parameters corresponding to the production line.

[0108] As a further implementation of the method, the method also includes:

[0109] Obtain the current energy consumption parameters of the production line and input them into the energy consumption level prediction model to obtain the current energy consumption level of the production line.

[0110] In the above implementation, in order to make predictions based on the energy consumption level prediction model, the current energy consumption parameters of the production line are obtained, and the current energy consumption parameters are input into the energy consumption level prediction model to obtain the current energy consumption level corresponding to the production line.

[0111] This application also discloses an energy consumption optimization system based on the Industrial Internet of Things.

[0112] refer to Figure 2 An energy consumption optimization system based on the Industrial Internet of Things (IIoT) includes a management platform, a sensor network platform, and an object platform that are connected in sequence via communication. The management platform is configured with:

[0113] The data acquisition module is used to acquire historical setting parameter data, historical energy consumption data, and energy consumption rating data of each processing equipment in the production line;

[0114] The model building module is used to build an energy consumption prediction model for the production line based on historical set parameter data and historical energy consumption data, and to build an energy consumption level prediction model for the production line based on historical energy consumption data and energy consumption rating data.

[0115] The energy consumption parameter determination module is used to obtain the target energy consumption level, set energy consumption constraints, call the energy consumption level prediction model, and search according to the gradient descent algorithm to obtain the target energy consumption parameter. The predicted energy consumption level of the target energy consumption parameter is consistent with the target energy consumption level.

[0116] The parameter setting module is used to obtain the target setting parameters of the target processing equipment, set the setting parameter constraints, call the energy consumption prediction model, and search based on the particle swarm optimization algorithm to determine the optimal setting parameters of non-target processing equipment. Among them, the energy consumption prediction parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters.

[0117] The overall framework of another application scenario of the energy consumption optimization method based on the Industrial Internet of Things in this application is as follows: Figure 3 As shown, it can include a user platform, service platform, management platform, sensor network platform, and object platform that interact sequentially, forming a five-platform architecture based on the Industrial Internet of Things (IIoT). Specifically, the management platform includes a target production rate determination module, an environmental control area division module, an environmental control area generation module, and an environmental control instruction generation module; the service platform includes a main service database, n service sub-platforms, and n service sub-databases. Each service sub-platform can communicate with its corresponding service sub-database, and each service sub-database can communicate with the main service database.

[0118] Specifically, in the aforementioned application scenario, the energy consumption optimization method based on the Industrial Internet of Things (IIoT) includes a management platform configured to: acquire historical setting parameter data, historical energy consumption data, and energy consumption rating data for each processing equipment in the production line; construct an energy consumption prediction model for the production line based on the historical setting parameter data and historical energy consumption data, and construct an energy consumption level prediction model for the production line based on the historical energy consumption data and energy consumption rating data; acquire the target energy consumption level, set energy consumption constraints, call the energy consumption level prediction model, and search using the gradient descent algorithm to obtain the target energy consumption parameters, wherein the energy consumption prediction level of the target energy consumption parameters is consistent with the target energy consumption level; acquire the target setting parameters for the target processing equipment, set setting parameter constraints, call the energy consumption prediction model, and search using the particle swarm optimization algorithm to determine the optimal setting parameters for non-target processing equipment, wherein the energy consumption prediction parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters.

[0119] By leveraging the interaction between various functional platforms in the industrial IoT-based energy consumption optimization method based on the aforementioned three or five platforms, a complete closed-loop information operation logic is established, ensuring the orderly operation of sensing and control information and realizing intelligent equipment management.

[0120] The energy consumption optimization system based on the Industrial Internet of Things of the present invention can implement any of the energy consumption optimization methods based on the Industrial Internet of Things, and the specific working process of the energy consumption optimization system based on the Industrial Internet of Things of the present invention can refer to the corresponding process in the above-mentioned energy consumption optimization methods based on the Industrial Internet of Things.

[0121] This application also discloses a computer device.

[0122] refer to Figure 4 A computer device includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement any of the above-described energy consumption optimization methods based on the Industrial Internet of Things.

[0123] This application also discloses a computer-readable storage medium.

[0124] A computer-readable storage medium storing a computer program that can be loaded by a processor and executed any of the above-described energy optimization methods based on the Industrial Internet of Things.

[0125] The computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0126] 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. An energy consumption optimization method based on the Industrial Internet of Things, characterized in that, Applied to an industrial Internet of Things (IIoT) system, the IIoT system includes a management platform, a sensor network platform, and an object platform that are sequentially and communicatively connected. The method is executed by the management platform and includes: Acquire historical setting parameter data, historical energy consumption data, and energy consumption rating data for each processing equipment in the production line; An energy consumption prediction model for the production line is constructed based on the historical set parameter data and the historical energy consumption data, and an energy consumption level prediction model for the production line is constructed based on the historical energy consumption data and the energy consumption rating data. Obtain the target energy consumption level, set energy consumption limits, call the energy consumption level prediction model, and search according to the gradient descent algorithm to obtain the target energy consumption parameters, wherein the energy consumption prediction level of the target energy consumption parameters is consistent with the target energy consumption level; Obtain the target setting parameters of the target processing equipment, set the setting parameter constraints, call the energy consumption prediction model, and search based on the particle swarm optimization algorithm to determine the optimal setting parameters of the non-target processing equipment. Among them, the energy consumption prediction parameters corresponding to the optimal setting parameters and the target setting parameters are closest to the target energy consumption parameters.

2. The energy consumption optimization method based on the Industrial Internet of Things according to claim 1, characterized in that, The step of constructing the energy consumption prediction model for the production line based on the historical set parameter data and the historical energy consumption data includes: The historical setting parameter data and the historical energy consumption data are used as the first model training data, and the first model training data is divided into a first training set and a first test set according to a first preset ratio. The hyperparameters of the pre-built decision tree model are set according to the random network search algorithm, and the first root mean square error (RMSE) and the first coefficient of determination (R²) are set. 2 As the primary evaluation indicator; The pre-built decision tree model is trained based on the first training set to obtain a trained decision tree model; The trained decision tree model is tested according to the first test set, and the corresponding error is determined according to the first evaluation index. If it is within the first preset range, the trained decision tree model is used as the energy consumption prediction model of the production line.

3. The energy consumption optimization method based on the Industrial Internet of Things according to claim 1, characterized in that, The step of constructing the energy consumption level prediction model for the production line based on the historical energy consumption data and the energy consumption rating data includes: The historical energy consumption data and the energy consumption rating data are used as training data for the second model, and the training data for the second model is divided into a second training set and a second test set according to a second preset ratio. The hyperparameters of the pre-built neural network model are set according to the random network search algorithm, and the second root mean square error (RMSE) and the second coefficient of determination (R²) are set. 2 As a second evaluation indicator; The pre-built neural network model is trained according to the second training set to obtain a trained neural network model; The trained neural network model is tested according to the second test set, and the corresponding error is determined according to the second evaluation index to be within the second preset range. If it is, the trained neural network model is used as the energy consumption level prediction model of the production line.

4. The energy consumption optimization method based on the Industrial Internet of Things according to claim 1, characterized in that, The step of calling the energy consumption level prediction model and searching according to the gradient descent algorithm to obtain the target energy consumption parameters includes: A first search space is set according to the energy consumption limit conditions, a first initial value is set according to the first search space, and a first initial search direction is set. Based on the iteration point and search direction The iteration is performed within the first search space, wherein... For the (k+1)th iteration point, For the k-th iteration point, For the search step size, The search direction at the k-th iteration point. The search direction at the (k+1)th iteration point. Let be the update factor for the k-th iteration point. The gradient at the (k+1)th iteration point; During the iteration process, for each iteration point, energy consumption level is predicted according to the energy consumption level prediction model, and it is determined whether the corresponding predicted energy consumption level is consistent with the target energy consumption level. If so, the corresponding iteration point is used as the target energy consumption parameter.

5. The energy consumption optimization method based on the Industrial Internet of Things according to claim 1, characterized in that, The step of calling the energy consumption prediction model and searching based on the particle swarm optimization algorithm to determine the optimal setting parameters for the non-target processing equipment includes: A second search space is set according to the set parameter constraints, and a set of particles is set according to the second search space and the target set parameters, and the velocity of each particle is set. Update formula based on location and speed update formula The process iterates over each particle within the second search space, where... For the (k+1)th iteration point, For the k-th iteration point, The velocity in the k-th iteration, Let w be the velocity in the (k+1)th iteration, and w be the inertia weight. For cognitive factors, A random number between 0 and 1 To be the best location found during the search process, As a social learning factor, A random number between 0 and 1 The best position found globally. For random number functions, The amplitude of the disturbance; During the iteration process, for each iteration of a group of particles, the energy consumption of each particle's iteration point is predicted according to the energy consumption prediction model, and it is determined whether the Pearson correlation coefficient between the corresponding first energy consumption prediction parameter and the target energy consumption parameter is less than a preset value. If so, the optimal setting parameters of the non-target processing equipment are determined according to the corresponding iteration point, and the iteration process ends. Otherwise, n parameter points are randomly selected in the second search space according to the target setting parameters, and the distance between the energy consumption prediction parameter corresponding to each parameter point and the target energy consumption parameter is calculated. The shortest distance is determined based on the distance, and the optimal setting parameters for the non-target processing equipment are determined based on the parameter point corresponding to the shortest distance.

6. The energy consumption optimization method based on the Industrial Internet of Things according to claim 1, characterized in that, The method further includes: Obtain the current setting parameters of the production line, wherein the current setting parameters include at least one or more of the following for each processing equipment: rotational speed, feed rate, cutting speed, vibration frequency, stress characteristics, pulse parameters, coolant flow rate, and coolant pressure; The current set parameters are input into the energy consumption prediction model to obtain the predicted value of the current energy consumption parameters corresponding to the production line.

7. The energy consumption optimization method based on the Industrial Internet of Things according to claim 1, characterized in that, The method further includes: The current energy consumption parameters of the production line are obtained and input into the energy consumption level prediction model to obtain the current energy consumption level corresponding to the production line.

8. An energy consumption optimization system based on the Industrial Internet of Things, characterized in that, It includes a management platform, a sensor network platform, and an object platform that are connected in sequence. The management platform is configured with: The data acquisition module is used to acquire historical setting parameter data, historical energy consumption data, and energy consumption rating data of each processing equipment in the production line; The model building module is used to build an energy consumption prediction model for the production line based on the historical set parameter data and the historical energy consumption data, and to build an energy consumption level prediction model for the production line based on the historical energy consumption data and the energy consumption rating data. The energy consumption parameter determination module is used to obtain the target energy consumption level, set energy consumption limits, call the energy consumption level prediction model, and search according to the gradient descent algorithm to obtain the target energy consumption parameter, wherein the energy consumption prediction level of the target energy consumption parameter is consistent with the target energy consumption level. The parameter setting module is used to obtain the target setting parameters of the target processing equipment, set the setting parameter constraints, call the energy consumption prediction model, and search based on the particle swarm optimization algorithm to determine the optimal setting parameters of the non-target processing equipment. The optimal setting parameters and the energy consumption prediction parameters corresponding to the target setting parameters are closest to the target energy consumption parameters.

9. A computer device, characterized in that, The method includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and execute the method of any one of claims 1 to 7.

Images