An intelligent power-saving controller system

Abstract

The present invention relates to the technical field of equipment power-saving control and relates to an intelligent power-saving controller system. The present invention generates real-time working state parameters of equipment by collecting spectral characteristic parameters and load fluctuation characteristics of multi-process equipment, and outputs a target production beat time sequence for the coordinated operation of multi-process equipment by constructing a time series neural network model, so as to realize dynamic prediction and coordinated planning of the production beat; at the same time, through the weighted analysis of the fusion of the historical operation response duration, the vibration spectrum characteristic deviation degree and the load fluctuation characteristic deviation degree, a time sequence of equipment operation interval durations is generated, and start-stop control instructions for each process equipment are dynamically allocated to realize the adaptive optimization of the production time sequence, and the no-load standby state is identified after control to realize the accurate determination and marking of ineffective energy consumption. The present invention provides a systematic solution for the intelligent power-saving control and energy efficiency optimization of multi-process equipment through data-driven decision-making and dynamic control closed-loop.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment power-saving control and relates to an intelligent power-saving controller system. Background Art

[0002] In the industrial production process, the energy consumption management of multi-process equipment operating in coordination is a key link to improve production efficiency and reduce costs. Traditional equipment control methods often rely on fixed operating parameters or simple threshold judgments, and it is difficult to perceive the actual working state changes of equipment in real time, resulting in problems such as uncoordinated production rhythms and excessive no-load standby time of equipment, causing energy waste and low production efficiency. Therefore, there is an urgent need for an intelligent power-saving control technology that can monitor the equipment status in real time, accurately predict the production rhythm, and dynamically adjust the operation strategy to achieve the efficient coordination and energy-saving operation of multi-process equipment.

[0003] For example, a power load minute-level flexible control method, system and equipment with the Chinese patent publication number CN118412864B constructs a precision-charge model of different industrial equipment in each production and processing link by simulating the working state information of each industrial equipment under different processing precisions; uses the precision-charge model to predict the charge demand distribution during the production task process and optimize the equipment scheduling of the production task. Although this method can reduce load fluctuations, its core still relies on preset charge thresholds and static process classifications, and does not solve problems such as beat conflicts and no-load energy consumption accumulation caused by real-time state deviations in the coordinated operation of multi-process equipment. In addition, this technology lacks analysis of the dynamic correlation between the vibration characteristics and current fluctuations of equipment, and it is difficult to achieve precise start-stop timing control.

[0004] However, the existing technologies have the following problems: 1. The existing technologies mainly focus on the minute-level optimal scheduling of power loads, and are insufficient in monitoring and accurately grasping the real-time working states of equipment during coordinated operation in different processes. It is difficult to dynamically adjust the production rhythm according to the actual operating conditions of the equipment, resulting in a low matching degree between the production rhythm prediction and the actual equipment operating state, easily causing chaotic rhythms in the coordinated operation between equipment, increasing the idling or waiting time of the equipment, reducing production efficiency and wasting energy.

[0005] 2. The existing technologies rely on the precision-charge model and do not conduct a fusion analysis of the response duration and beat timing between process equipment. It is easy to cause a sharp increase in overall energy consumption due to the delay of a single equipment, resulting in the inability to dynamically optimize the start-up time of multi-process equipment in a timely manner, thereby reducing the coordinated efficiency between processes and exacerbating the power load fluctuations. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent power-saving controller system to solve the problems existing in the above-mentioned existing technologies and achieve precise power-saving control of multi-process equipment.

[0007] The technical solution adopted by the present invention to solve its technical problems is as follows: An intelligent power-saving controller system includes a working state generation module, which is used to collect the vibration time series signal and current time series signal of multi-process equipment in real time, extract the spectral feature parameters in the vibration time series signal and the load fluctuation feature of the current time series signal, and generate the real-time working state parameters of the equipment.

[0008] A production beat prediction module, which is used to input the real-time working state parameters of the equipment into the production beat prediction model preset for the corresponding equipment, and output the target production beat time series sequence when the multi-process equipment operates in coordination.

[0009] A dynamic adjustment instruction generation module, which is used to extract the historical operation response duration of the multi-process equipment, fuse and analyze the data of the time series deviation between it and the real-time working state parameters of the equipment, obtain the equipment operation interval duration time series sequence, dynamically allocate the start time control instructions of each process equipment based on the equipment operation interval duration time series sequence, and send the instructions to the controller.

[0010] An unloaded state recognition module, which is used to monitor the vibration time series signal and current time series signal of the multi-process equipment after control, and determine the unloaded standby state of the equipment based on the preset unloaded standby state determination rule.

[0011] An unloaded state processing module, which is used to execute control operations through the controller according to the duration of the unloaded standby state of the equipment.

[0012] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention uses the working state generation module to collect the vibration time series signal and current time series signal of multi-process equipment in real time, and extracts the characteristic parameters to generate the real-time working state parameters of the equipment, realizing the dynamic and accurate recognition of the equipment working state, and providing a reliable data basis for subsequent production beat prediction and energy-saving control.

[0013] (2) The present invention uses a time series neural network to construct a production beat prediction model, predicts the target production beat time series sequence based on the real-time working state parameters of the equipment, improves the accuracy of production beat prediction, enhances the beat matching degree when the multi-process equipment operates in coordination, and reduces the idling or waiting time of the equipment caused by inconsistent beats.

[0014] (3) The present invention fuses the historical operation response duration and the time series deviation of the real-time working state parameters, dynamically allocates the start time control instructions of each process equipment, makes the allocation of equipment start time more in line with the actual operation state, reduces the waiting loss between equipment, shortens the production cycle, and reduces the energy consumption of the equipment in the non-essential operation stage.

[0015] (4) Based on a preset determination rule, the present invention monitors the no-load standby state of the device and performs a shutdown control operation according to the duration of no-load standby, achieving timely and accurate determination of the no-load standby state of the device and energy-saving control, further reducing the power consumption of the device during no-load standby and improving the power-saving effect of the entire multi-process device system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0017] Figure 1 It is a schematic diagram of the connection of the system modules of the present invention.

[0018] Figure 2 It is a schematic diagram of the model training process of the temporal neural network in the present invention.

[0019] Figure 3 It is a schematic diagram of the steps for obtaining the time series of the standby operation interval duration in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] Now, various exemplary embodiments of the present invention will be described in detail with reference to the drawings. It should be noted that: Unless otherwise specifically stated, the relative arrangements, numerical expressions, and numerical values of the components and steps described in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that for the sake of description, the dimensions of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0021] The following description of at least one exemplary embodiment is actually only illustrative and in no way limits the present invention and its application or use. Technologies, methods, and devices known to those of ordinary skill in the relevant field may not be discussed in detail, but where appropriate, the said technologies, methods, and devices should be regarded as part of the specification.

[0022] In all the examples shown and discussed here, any specific value should be interpreted as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0023] Please refer to Figure 1As shown in the figure, the present invention provides an intelligent power-saving controller system, including a working state generation module, a production rhythm prediction module, a dynamic adjustment instruction generation module, an idle state recognition module, and an idle state processing module. The connection relationship between the modules is as follows: the working state generation module is connected to the production rhythm prediction module, the dynamic adjustment instruction generation module is respectively connected to the production rhythm prediction module and the idle state recognition module, and the idle state recognition module is connected to the idle state processing module.

[0024] The working state generation module is used to collect the vibration time series signal and current time series signal of multi-process equipment in real time, extract the spectral feature parameters in the vibration time series signal and the load fluctuation feature of the current time series signal, and generate the real-time working state parameters of the equipment.

[0025] Among them, the multi-process equipment undertakes different tasks in the industrial production process, and there is a certain association and cooperation between them. The working state generation module is responsible for collecting the vibration time series signal and current time series signal of the multi-process equipment in real time. For example, in an automobile parts production line, the equipment of different processes such as stamping, welding, and painting all have their specific vibration and current change patterns, and these signals can reflect the current operating condition of the equipment.

[0026] Vibration sensors are generally installed at key parts of the equipment, such as bearings, rotating shafts, etc., and can sense the vibration time series signal generated during the operation of the equipment. The current sensor is connected to the power supply line of the equipment to monitor the current time series signal flowing through the equipment in real time.

[0027] For the vibration time series signal, signal processing technology is used for spectral analysis to extract the spectral feature parameters therein. The spectral feature parameters include but are not limited to the main frequency, frequency bandwidth, harmonic content, etc. The main frequency reflects the vibration frequency of the main vibration source of the equipment, the frequency bandwidth reflects the frequency range covered by the vibration signal, and the harmonic content can indicate whether there are abnormalities such as imbalance and looseness in the equipment. For example, when the main bearing of the equipment wears, the spectral characteristics of the vibration signal will change, the main frequency may shift, and the harmonic content will also increase accordingly.

[0028] Analyze the load fluctuation feature of the current time series signal, extract the current at each time point in the current time series signal, and calculate and compare to obtain the current mean value and current peak value. At the same time, methods such as wavelet analysis are used to decompose and reconstruct the current signal to extract the current fluctuation period in different frequency bands. For example, during the start-up and stop processes of the equipment, the current signal will have large fluctuations, and at this time, the current fluctuation period will be significantly different from the state when the equipment is running stably.

[0029] Among them, the average current reflects the average power consumption of the device over a period of time. The average current of the device will be different under different working conditions. For example, the average current will be relatively large when the device is running under heavy load; while the average current will be small when the device is running under light load or no load. The current peak represents the maximum value of the current during the operation of the device. The magnitude of the current peak is related to the startup of the device, load mutation, etc. For example, when the device starts up, due to the large starting current of the motor, an obvious current peak will appear. The current fluctuation period reflects the variation law of the current with time. The working process of some devices is periodic, and their current will also show corresponding periodic fluctuations.

[0030] The present invention uses a working state generation module to collect the vibration time series signal and current time series signal of a multi-process device in real time, extract characteristic parameters to generate the real-time working state parameters of the device, and realize the dynamic and accurate identification of the device working state, providing a reliable data basis for subsequent production beat prediction and energy-saving control.

[0031] The production beat prediction module is used to input the real-time working state parameters of the device into the production beat prediction model preset for the corresponding device, and output the target production beat time series sequence when the multi-process device operates in coordination. Among them, the target production beat time series sequence is generated by arranging the production beat times of the multi-process device in ascending order of time sequence.

[0032] It should be noted that the construction method of the production beat prediction model is as follows: extract the working state parameters and corresponding production beat data of the multi-process device from the historical production data, and perform time sequence alignment on the working state parameters and production beat data.

[0033] Taking the main frequency, frequency bandwidth, harmonic content in the vibration spectrum characteristic parameters and the average current, peak value and fluctuation period in the load fluctuation characteristic as input features, and the production beat data as the output label, the model is trained through a time sequence neural network to obtain the production beat prediction model of the multi-process device.

[0034] Furthermore, the historical production data is stored in the database or log file through the operation records of the multi-process device during the process of the factory producing the same product. For example, an automobile manufacturing factory collects the production data of each process device such as stamping, welding, and painting in the past year, covering the data under the production of different types of vehicle models, and cleans the collected historical production data to remove obvious error and abnormal data points, such as extremely large or extremely small values in the current signal due to sensor failures, and then performs feature extraction and selection on the cleaned data to obtain the working state parameters of the multi-process device and the corresponding production beat data.

[0035] The working state parameters include vibration spectrum characteristic parameters, average current, peak value, and fluctuation period parameters, etc. The vibration spectrum characteristic parameters reflect the frequency distribution of the equipment vibration; the average current, peak value, and fluctuation period parameters reflect the power consumption characteristics of the equipment.

[0036] The production cycle data refers to the time required for multi-process equipment to continuously complete a product or a process, which is an important indicator for measuring production efficiency.

[0037] Since the working state parameters and production cycle data may be recorded at different time scales or sampling frequencies, time series alignment is required. For example, the vibration sensor may collect data at a higher frequency, while the production cycle data is recorded once for each completed product. Align the working state parameters and production cycle data in time to ensure the consistency of input features and output labels in time.

[0038] As Figure 2 shown, the model training process of the time series neural network includes: constructing a time series training data set based on the time series-aligned working state parameters and production cycle data, and performing data preprocessing on the time series training data set, including normalization, missing value imputation, and outlier removal, to generate a preprocessed data set.

[0039] Based on the preprocessed data set, select a set time series neural network for model construction, adjust the hyperparameters of the model through cross-validation, and repeatedly optimize and determine the final model structure. The set time series neural network can be a long short-term memory network (LSTM). This neural network is an existing technology and will not be elaborated here.

[0040] Divide the preprocessed data set into a training set and a validation set according to a predetermined ratio, and perform iterative training on the training set through the final model structure to obtain the mapping relationship between the vibration spectrum characteristics and load fluctuation characteristics and the production cycle data. The predetermined ratio can be .

[0041] Substitute the working state parameters of the validation set into the obtained mapping relationship for evaluation and determination. If the determination result is the same as the production cycle data of the validation set, it indicates that the model training is successful, and then output the production cycle prediction model.

[0042] Furthermore, the cross-validation method is adopted to divide the preprocessed data set into multiple training-validation subsets, and through repeated training and validation, systematically adjust the hyperparameters of the model. The hyperparameters include learning rate, number of network layers, and number of neurons, so as to avoid the risk of overfitting and select a model structure with better generalization ability. During the training process, continuously monitor the accuracy rate, and according to the fluctuation range of the accuracy rate, finely adjust the model parameters to lay a solid foundation for the production cycle prediction model.

[0043] The present invention uses a time series neural network to construct a production beat prediction model, predicts the target production beat time series based on the real-time working state parameters of the equipment, improves the accuracy of production beat prediction, enhances the beat matching degree during the collaborative operation of multi-process equipment, and reduces the idling or waiting time of the equipment caused by inconsistent beats.

[0044] The dynamic adjustment instruction generation module is used to extract the historical operation response duration of multi-process equipment, perform fusion data analysis on the time series deviation between it and the real-time working state parameters of the equipment, obtain the equipment operation interval duration time series, dynamically allocate the start time control instructions for each process equipment based on the equipment operation interval duration time series, and send the instructions to the controller.

[0045] As Figure 3 shown, the method for obtaining the equipment operation interval duration time series is as follows: S1. Extract the response duration of multi-process equipment in each historical operation from the historical production data record, and analyze the historical operation response delay degree based on the response duration of each historical operation.

[0046] Further, the process of analyzing the historical operation response delay degree is as follows: Calculate the mean value of the response duration of each historical operation to obtain the mean response duration, obtain the difference between the response duration of each historical operation and the mean response duration, calculate the ratio of the difference to the mean response duration, and take the average value of the ratio calculation results as the historical operation response delay degree.

[0047] S2. Obtain the vibration spectrum feature and load fluctuation feature from the real-time working state parameters of the multi-process equipment, calculate the time series deviation between the vibration spectrum feature and the load fluctuation feature and the reference vibration spectrum feature threshold and the reference load fluctuation feature threshold of the corresponding equipment's standardized working state parameters respectively, obtain the vibration spectrum feature deviation degree and the load fluctuation feature deviation degree, and perform weighted fusion analysis on the historical operation response delay degree, the vibration spectrum feature deviation degree, and the load fluctuation feature deviation degree to obtain the influence degree of the equipment operation interval duration of the multi-process equipment.

[0048] Further, the calculation formula for the vibration spectrum feature deviation degree is , where is the vibration spectrum feature deviation degree, is the corresponding value of the vibration spectrum feature, is the reference vibration spectrum feature threshold; similarly, the load fluctuation feature deviation degree is obtained by using the calculation method of the vibration spectrum feature deviation degree.

[0049] The analysis method for the influence degree of the equipment operation interval duration of the multi-process equipment is , where is the load fluctuation feature deviation degree, is the historical operation response delay degree, is the influence weight corresponding to the deviation degree of the vibration spectrum characteristics, reflecting the influence degree of the abnormal vibration spectrum of the equipment on the operation interval duration. For example, a large deviation in the vibration spectrum characteristics may indicate the risk of mechanical failure, and it is necessary to increase the equipment operation interval duration. is the influence weight corresponding to the deviation degree of the load fluctuation characteristics, representing the influence weight of the load fluctuation on the stability of the collaborative operation of the equipment. When the load fluctuates violently, it is necessary to adjust the equipment operation interval duration to balance energy consumption and efficiency. is the influence weight corresponding to the historical operation response delay degree, reflecting the reference value of the historical response delay of the equipment for the current operation interval planning. Equipment with frequent response delays needs to reserve more buffer time. Usually, it satisfies the normalization condition: , and the value range of a single weight is [0, 1].

[0050] In a specific embodiment, such as in a certain automobile assembly line, due to frequent load fluctuations, such as = 0.6, the system preferentially adjusts the start-stop timing to balance energy consumption; at the same time, for the vibration sensitivity of precision welding equipment, such as = 0.3, shorten the detection interval; the weight of the historical response delay is relatively low, such as = 0.1, because the response stability of the equipment is relatively high. Through dynamic weight allocation, the calculation of the comprehensive influence degree σ is more in line with the actual needs, optimizing the production rhythm and energy efficiency.

[0051] S3. Take the combination result of the minimum operation interval duration of the multi-process equipment and the influence degree of the corresponding equipment operation interval duration as the current equipment operation interval duration, and arrange the current equipment operation interval durations of the multi-process equipment in sequence according to the time sequence to generate an equipment operation interval duration time sequence.

[0052] Furthermore, the analysis method of the current equipment operation interval duration is as follows: , where is the current equipment operation interval duration, is the minimum operation interval duration. An increase in the historical operation response delay degree indicates a decrease in the equipment response speed, and more time needs to be reserved to ensure the effective execution of instructions. Therefore, the equipment operation interval duration increases to avoid instruction conflicts or production rhythm breaks caused by slow equipment response; an increase in the deviation degree of the vibration spectrum characteristics indicates that the equipment has mechanical abnormalities, such as bearing wear and imbalance, and the interval needs to be extended to reduce the interference of abnormal vibration on production. Therefore, the equipment operation interval duration increases to facilitate intervention and maintenance; when the deviation degree of the load fluctuation characteristics is too large, the equipment needs a longer time to adjust its operation state to maintain stable output. Therefore, the equipment operation interval duration increases to set aside a buffer time to suppress the fluctuation.

[0053] It should be noted that the control instruction for dynamically allocating the start time of each process equipment specifically includes: comparing the current equipment operation interval duration of the multi-process equipment in the equipment operation interval duration time series with the corresponding equipment production transfer interval duration, and screening the maximum duration as the final operation interval duration of the multi-process equipment.

[0054] Combining the target production beat time series with the final operation interval duration of the multi-process equipment to obtain the start time of the multi-process equipment, and generating a start time control instruction according to the start time of the multi-process equipment.

[0055] It should be noted that the equipment production transfer interval duration refers to the shortest time interval required for material transfer after completing a process in industrial production. This time interval is determined based on the production beat of the equipment and the logistics requirements on the production line to ensure that materials can be transferred to the next process in a timely manner without affecting the normal operation of the equipment and the continuity of the production line.

[0056] Furthermore, the method for obtaining the start time of the multi-process equipment is as follows: extracting the production beat time of the multi-process equipment in the target production beat time series, combining the production beat time of the multi-process equipment with the final operation interval duration to obtain the production beat start and stop time of the multi-process equipment. When the final operation interval duration of the equipment is the current equipment operation interval duration, the start time of the equipment is the production beat stop time of the previous adjacent equipment corresponding to the equipment. When the final operation interval duration of the equipment is the equipment production transfer interval duration, the start time of the equipment is the production beat start time of the equipment minus the current equipment operation interval duration.

[0057] In a specific embodiment, for example, an automobile assembly line includes 3 key processes: Process A (body welding): equipment production transfer interval duration = 5 minutes, current equipment operation interval duration = 4 minutes.

[0058] Process B (painting and drying): equipment production transfer interval duration = 6 minutes, current equipment operation interval duration = 7 minutes.

[0059] Process C (final assembly and debugging): equipment production transfer interval duration = 8 minutes, current equipment operation interval duration = 6 minutes.

[0060] Target production beat time series (Process A, Process B, Process C) = (5 minutes, 7 minutes, 10 minutes).

[0061] The final operation interval duration is the maximum of the current device operation interval duration and the production transfer interval duration. Therefore, the final operation interval duration of Process A is: max(4, 5) = 5 minutes; the final operation interval duration of Process B is: max(7, 6) = 7 minutes; the final operation interval duration of Process C is: max(6, 8) = 8 minutes.

[0062] If the current time is 8:00, then the production beat start time of Process A is 8:00, and the production beat stop time is 8:05; the production beat start time of Process B is 8:12, and the production beat stop time is 8:19; the production beat start time of Process C is 8:27, and the production beat stop time is 8:37.

[0063] The start time of Process A: The final duration is the production transfer interval duration of the device, which is 5 minutes. The start time = the production beat start time of Process A, 08:00 - the current device operation interval duration, 4 minutes = 07:56.

[0064] The start time of Process B: The final duration is the current device operation interval duration, which is 7 minutes. The start time = the stop time of Process A, 08:05.

[0065] The start time of Process C: The final duration is the production transfer interval duration of the device, which is 8 minutes. The start time = the production beat start time of Process C, 8:27 - the current device operation interval duration, 6 minutes = 8:21.

[0066] The present invention combines the historical operation response duration and the timing deviation of real-time working state parameters, dynamically allocates the start time control instructions of each process device, makes the allocation of device start time more in line with the actual operation state, reduces the waiting loss between devices, shortens the production cycle, and reduces the energy consumption of the device in the non-essential operation stage.

[0067] The no-load state recognition module is used to monitor the vibration timing signal and current timing signal of the multi-process device after control, and determine the no-load standby state of the device based on the preset no-load standby state determination rule.

[0068] It should be noted that the preset no-load standby state determination rule is: If the current mean value in the current timing signal for a continuously set number of windows is within the current reference range of the device in the no-load standby state and the main frequency in the vibration timing signal is lower than the preset vibration frequency threshold, it is determined that the device is in the no-load standby state.

[0069] Further, the current reference range of the device in the no-load standby state is determined by extracting the current data of all no-load standby periods from the device's historical operation log, determining the normal distribution interval through statistical analysis, and using it as the reference range to avoid interference from occasional noise. The preset vibration frequency threshold refers to the vibration safety standard of the corresponding device, such as ISO 10816.

[0070] The no-load state processing module is used to execute control operations through the controller according to the duration of the no-load standby state of the device.

[0071] It should be noted that the specific content of the no-load state processing module is: real-time monitoring of the duration of the no-load standby state of the no-load standby device. If the duration of the no-load standby state is greater than the set standby time threshold, the controller will execute a shutdown control operation on the no-load standby device.

[0072] Based on the preset determination rule, the present invention monitors the no-load standby state of the device and executes a shutdown control operation according to the duration of the no-load standby, realizing the timely and accurate determination of the no-load standby state of the device and energy-saving control, further reducing the power consumption of the device during no-load standby, and improving the power-saving effect of the entire multi-process device system.

[0073] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0074] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0075] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0076] In addition, in each embodiment of the present application, the various functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0077] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the said claims.

[0078] Finally, the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent power-saving controller system, characterized in that, Including: A working state generation module, which is used to collect the vibration time series signal and current time series signal of multi-process equipment in real time, extract the spectral feature parameters in the vibration time series signal and the load fluctuation feature of the current time series signal, and generate the real-time working state parameters of the equipment; A production beat prediction module, which is used to input the real-time working state parameters of the equipment into the production beat prediction model of the preset corresponding equipment, and output the target production beat time series sequence when the multi-process equipment operates in coordination; A dynamic adjustment instruction generation module, which is used to extract the historical operation response duration of the multi-process equipment, fuse and analyze the data of the time series deviation between it and the real-time working state parameters of the equipment to obtain the time series sequence of the equipment operation interval duration, and dynamically allocate the start time control instructions of each process equipment based on the time series sequence of the equipment operation interval duration, and send the instructions to the controller; An unloaded state recognition module, which is used to monitor the vibration time series signal and current time series signal of the multi-process equipment after control, and determine the unloaded standby state of the equipment based on the preset unloaded standby state determination rule; An unloaded state processing module, which is used to execute control operations through the controller according to the duration of the unloaded standby state of the equipment; The construction method of the production beat prediction model is as follows: Extract the working state parameters and corresponding production beat data of the multi-process equipment from the historical production data, and perform time series alignment on the working state parameters and production beat data; Take the main frequency, frequency bandwidth, harmonic content in the vibration spectrum feature parameters and the current mean value, peak value and fluctuation period in the load fluctuation feature as input features, and the production beat data as the output label, and perform model training through a time series neural network to obtain the production beat prediction model of the multi-process equipment; The model training process of the time series neural network includes: Construct a time series training dataset according to the time series aligned working state parameters and production beat data, and perform data preprocessing on the time series training dataset, including normalization processing, missing value imputation and outlier removal, to generate a preprocessed dataset; Select a set time series neural network for model construction based on the preprocessed dataset, adjust the hyperparameters of the model through cross-validation, and repeatedly optimize and determine the final model structure; Divide the preprocessed dataset into a training set and a validation set according to a predetermined ratio, and perform iterative training on the training set through the final model structure to obtain the mapping relationship between the vibration spectrum features and load fluctuation features and the production beat data; Substitute the working state parameters of the validation set into the obtained mapping relationship for evaluation and determination. If the determination result is the same as the production beat data of the validation set, it indicates that the model training is successful, and then output the production beat prediction model.

2. The intelligent power-saving controller system according to claim 1, wherein: The obtaining method of the time series sequence of the equipment operation interval duration is as follows: Extract the response duration of the multi-process equipment in each historical operation from the historical production data record, and analyze the historical operation response delay degree based on the response duration of each historical operation; Obtain the vibration spectrum characteristics and load fluctuation characteristics from the real-time working state parameters of the multi-process equipment. Calculate the time-series deviation between the vibration spectrum characteristics and load fluctuation characteristics and the reference vibration spectrum characteristic threshold and reference load fluctuation characteristic threshold of the corresponding equipment's standardized working state parameters respectively, to obtain the vibration spectrum characteristic deviation degree and load fluctuation characteristic deviation degree. Conduct weighted fusion analysis on the historical operation response delay degree, vibration spectrum characteristic deviation degree, and load fluctuation characteristic deviation degree to obtain the influence degree of the equipment operation interval duration of the multi-process equipment. Use the combination result of the minimum operation interval duration of the multi-process equipment and the influence degree of the corresponding equipment operation interval duration as the current equipment operation interval duration, and generate an equipment operation interval duration time-series sequence.

3. The intelligent power-saving controller system according to claim 2, wherein: The process of analyzing the historical operation response delay degree is as follows: Calculate the mean value of the response durations of each historical operation to obtain the mean response duration. Obtain the difference between the response duration of each historical operation and the mean response duration, calculate the ratio of this difference to the mean response duration, and take the average value of the ratio calculation results as the historical operation response delay degree.

4. The intelligent power-saving controller system according to claim 2, wherein: The generation method of the equipment operation interval duration time-series sequence is to arrange the current equipment operation interval durations of the multi-process equipment in sequence according to the time series.

5. An intelligent power-saving controller system according to claim 1, characterized in that: The specific content of dynamically allocating the start time control instructions for each process equipment includes: Compare the current equipment operation interval duration of the multi-process equipment in the equipment operation interval duration time-series sequence with the corresponding equipment production transfer interval duration, and select the maximum duration as the final operation interval duration of the multi-process equipment. Combine the target production beat time-series sequence with the final operation interval duration of the multi-process equipment to obtain the start time of the multi-process equipment, and generate a start time control instruction based on the start time of the multi-process equipment. The equipment production transfer interval duration refers to the shortest time interval required for material transfer after completing one process in the industrial production process.

6. The intelligent power-saving controller system according to claim 5, wherein: The method for obtaining the start time of the multi-process equipment is as follows: Extract the production beat time of the multi-process equipment in the target production beat time-series sequence, combine the production beat time of the multi-process equipment with the final operation interval duration to obtain the production beat start and stop time of the multi-process equipment. When the final operation interval duration of the equipment is the current equipment operation interval duration, the start time of the equipment is the production beat stop time of the corresponding previous adjacent equipment. When the final operation interval duration of the equipment is the equipment production transfer interval duration, the start time of the equipment is the production beat start time minus the current equipment operation interval duration.

7. An intelligent power-saving controller system according to claim 1, characterized in that: The preset no-load standby state determination rule is: If the current mean values in a continuous set number of windows in the current time-series signal are all within the current reference range of the equipment in the no-load standby state and the main frequency in the vibration time-series signal is lower than the preset vibration frequency threshold, it is determined that the equipment is in the no-load standby state.

8. An intelligent power-saving controller system according to claim 1, characterized in that: The specific content of the no-load state processing module is: Real-time monitor the continuous duration of the no-load standby state of the no-load standby equipment. If the continuous duration of the no-load standby state is greater than the set standby time threshold, the controller performs a shutdown control operation on the no-load standby equipment.

Images