A method and system for evaluating power consumption of factory workshops

Abstract

The present invention discloses a method and system for evaluating the power energy consumption of a factory workshop, and relates to the technical field of energy consumption evaluation and diagnosis, including obtaining basic workshop data and performing preprocessing; performing feature extraction and combination to generate a multivariate feature sample set; constructing an equipment energy consumption prediction model based on the multivariate feature sample set to obtain a predicted energy consumption value; calculating a real-time predicted energy consumption value, and performing deviation value calculation to obtain an evaluation result; the method of the present invention forms a comprehensive multivariate feature sample set through deep feature mining and combination, and then constructs an equipment energy consumption prediction model. This process not only improves the prediction accuracy, but also can monitor the equipment energy consumption in real time and quickly evaluate the deviation, thereby effectively optimizing the energy management in the production process, achieving the goal of energy conservation and emission reduction, and promoting environmental sustainable development while improving the economic benefits of the enterprise.

Description

Technical Field

[0001] The present invention relates to the technical field of energy consumption evaluation and diagnosis, and in particular to a method and system for evaluating the power energy consumption of a factory workshop. Background Art

[0002] Under the trend of Industry 4.0, smart factories have become the key to improving efficiency and sustainability, among which power consumption assessment is particularly critical, directly affecting resource optimization and environmental impact. The traditional energy consumption management method that relies on manual records is difficult to achieve real-time and accurate monitoring and the discovery of potential energy-saving opportunities. Although data is automatically collected through sensors, it is still limited to data aggregation and lacks in-depth analysis capabilities. These systems have significant shortcomings in energy consumption prediction accuracy, multi-factor comprehensive analysis, and real-time feedback deviation adjustment mechanisms. They ignore the complex impact of production environment fluctuations, equipment aging, etc. on energy consumption, and fail to achieve dynamic optimization of energy consumption management. Therefore, current technologies are limited in providing detailed energy consumption assessments and immediate energy consumption improvement recommendations, and more intelligent and comprehensive solutions are urgently needed. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides a method and system for evaluating power consumption in a factory workshop to solve the problem of difficulty in achieving real-time accurate monitoring and discovering potential energy-saving opportunities.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a method for evaluating power consumption of a factory workshop, which includes installing a variety of sensors, acquiring basic workshop data, and performing preprocessing;

[0007] Based on the preprocessed workshop basic data, feature combination is performed to generate a multivariate feature sample set;

[0008] Based on the multivariate feature sample set, the equipment energy consumption prediction model is constructed to obtain the predicted energy consumption value;

[0009] Input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value;

[0010] Calculate the deviation between the real-time predicted energy consumption value and the actual energy consumption to obtain the evaluation result, thus realizing the evaluation of the power energy consumption of the factory workshop;

[0011] The basic data of the workshop include current, voltage, power factor, active power, reactive power, temperature, humidity, light intensity, air quality, vibration, rotation speed, pressure and flow.

[0012] As a preferred solution of the method for evaluating power energy consumption of a factory workshop of the present invention, the multiple sensors include power metering sensors, environmental sensors and equipment status monitoring sensors.

[0013] As a preferred solution of the method for evaluating the power consumption of a factory workshop of the present invention, the preprocessing includes:

[0014] First, define the basic data of the workshop as d i , for d i Perform minimum and maximum scaling to obtain

[0015] Then, Gaussian filtering is applied to remove noise. After filtering, we get

[0016]

[0017] Where σ is the standard deviation of the Gaussian kernel, k represents the position of the current data point, j is the index within the window, ∑ j Refers to all d in the window ij Perform the summation.

[0018] As a preferred solution of the method for evaluating power consumption of a factory workshop of the present invention, the feature combination is performed to generate a multivariate feature sample set, including:

[0019] Define a weighted feature enhancement function H i , where i = 1, 2, ..., n, expressed as,

[0020]

[0021] In the formula, represents the Uth filtered eigenvalue, Γ is the gamma function, α represents the shape parameter, β represents the scaling parameter, γ represents the offset parameter, and w i represents the adaptive weight, Where η is a parameter that controls the smoothness of weight distribution, n is the number of features, represents the variance of the jth feature, represents the variance of the i-th feature;

[0022] Define a feature combination function C to enhance the feature function H i Combine features to obtain a multivariate feature sample set V k , where k represents the number of samples, expressed as,

[0023]

[0024] In the formula, is the kernel function, θ k Represents the parameters in the kernel function.

[0025] As a preferred solution of the method for evaluating the power consumption of a factory workshop of the present invention, the equipment energy consumption prediction model is constructed to obtain the predicted energy consumption value including:

[0026] Based on the multivariate feature sample set V k , build the equipment energy consumption prediction model, expressed as,

[0027]

[0028] In the formula, E pred Represents the predicted energy consumption value, V k =[V k1 , V k2 , ..., V kj ] represents the multivariate feature vector of the kth sample, where V kj is the value of the jth feature, j represents the total number of features, F and F -1 denote Fourier transform and inverse Fourier transform respectively, w kj represents the Fourier coefficient, N NN (W,V k ) represents the output of the neural network model, W is the set of network parameters, and k is the number of samples.

[0029] As a preferred solution of the method for evaluating the power consumption of a factory workshop of the present invention, wherein: inputting the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value includes collecting real-time data d ti , where t represents the time point, and the real-time multivariate feature sample set is obtained through preprocessing and feature combination formula Real-time multivariate feature sample set Input the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value E pred，t .

[0030] As a preferred solution of the method for evaluating the power consumption of a factory workshop of the present invention, the deviation between the real-time predicted energy consumption value and the actual energy consumption is calculated to obtain the evaluation result including:

[0031] The energy consumption value E will be predicted in real time pred，t The actual energy consumption value E at the same time step actual，t Compare and calculate the deviation value ΔE t , expressed as,

[0032] ΔE t =E actual,t -E pred,t ;

[0033] Set the deviation threshold to A;

[0034] When ΔE t When ≤A, it indicates that the equipment energy consumption prediction model is highly consistent with the actual energy consumption, and the equipment energy consumption prediction model is used;

[0035] When ΔE t When >A, it indicates that the prediction deviates greatly from the actual situation. By adjusting the standard deviation σ of the kernel function, adjusting parameters α, β, γ, and adaptive weight w i , kernel function parameter θ k , optimize the equipment energy consumption prediction model until ΔE t ≤A.

[0036] In a second aspect, the present invention provides a factory workshop power energy consumption evaluation system, including a data acquisition and preprocessing module, a feature engineering module, an energy consumption prediction model building module, a real-time energy consumption prediction module and an energy consumption evaluation module.

[0037] Specifically, the data acquisition and preprocessing module is used to obtain basic workshop data and preprocess the basic workshop data;

[0038] The feature engineering module is mainly used to combine the features of the pre-processed workshop basic data to generate a multivariate feature sample set;

[0039] The energy consumption prediction model building module is used to build a device energy consumption prediction model based on a multivariate feature sample set to obtain a predicted energy consumption value;

[0040] The real-time energy consumption prediction module is used to input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value;

[0041] The energy consumption evaluation module is used to calculate the deviation between the real-time predicted energy consumption value and the actual energy consumption to obtain the evaluation result.

[0042] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for evaluating power consumption of a factory workshop as described in the first aspect of the present invention is implemented.

[0043] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for evaluating power consumption of a factory workshop as described in the first aspect of the present invention is implemented.

[0044] The beneficial effects of the present invention are as follows: for the evaluation of power energy consumption in factory workshops, comprehensive data is collected using multiple types of sensors, and data quality is improved through preprocessing; innovative algorithms are used to strengthen and combine features, reduce dimensions while retaining important information, and establish a prediction model that combines Fourier transform and neural network to accurately predict energy consumption; real-time data is input into the model to calculate the deviation value, and the prediction accuracy is evaluated with the mean square error to guide energy consumption management and strategy adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0046] Figure 1 This is a flow chart of the method for evaluating power consumption of a factory workshop in Example 1. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0050] Example 1, reference Figure 1 , which is the first embodiment of the present invention, provides a method for evaluating power consumption of a factory workshop, comprising the following steps:

[0051] S1. Install various sensors to obtain basic workshop data and perform preprocessing.

[0052] A variety of sensors including electricity metering sensors, environmental sensors, and equipment status monitoring sensors;

[0053] The basic data of the workshop include current, voltage, power factor, active power, reactive power, temperature, humidity, light intensity, air quality, vibration, speed, pressure, and flow;

[0054] First, define the basic data of the workshop as d i , for d i Perform minimum and maximum scaling to obtain It is expressed as,

[0055]

[0056] In the formula, represents the value of the i-th feature after preprocessing, d i Indicates the definition of basic workshop data, min(d i ) and max(d i ) represent the minimum and maximum values ​​of the i-th feature in the entire data set respectively;

[0057] Then, Gaussian filtering is applied to remove noise. After filtering, we get

[0058]

[0059] Where σ is the standard deviation of the Gaussian kernel, k represents the position of the current data point, j is the index within the window, ∑ j Refers to all d in the window ij Perform summation;

[0060] It should be noted that the preprocessing step ensures the consistency of the feature scales through minimum-maximum normalization, and applies Gaussian filtering to effectively suppress noise, which improves the data quality and the stability of model training, and provides an accurate and standardized basis for subsequent feature analysis and energy consumption prediction.

[0061] S2. Based on the preprocessed workshop basic data, feature combination is performed to generate a multivariate feature sample set.

[0062] The Pearson correlation coefficient is used to calculate the correlation between features and construct an n×n correlation coefficient matrix, where each element r i Represents each feature The degree of correlation between

[0063] Set the correlation threshold to 0.9;

[0064] When i When it is less than 0.9, it indicates low correlation, indicating that there is no redundancy between features;

[0065] When iWhen the value is greater than or equal to 0.9, it indicates a high correlation, indicating that there may be redundancy between features. Check the correlation coefficient matrix, find out the feature pairs with correlation coefficients higher than the threshold, and remove the features in each pair that have little contribution to the model explanation or have no obvious business significance;

[0066] For the features of nonlinear relationships, mutual information is used as a supplementary analysis method;

[0067] It should be noted that mutual information can measure the degree of mutual dependence of two variables and is suitable for detecting nonlinear correlation. The mutual information of each pair of features is calculated and compared with a pre-set threshold to further screen features;

[0068] Define a weighted feature enhancement function H i , where i = 1, 2, ..., n, the expression is:

[0069]

[0070] In the formula, represents the eigenvalue after the i-th filter, Γ is the gamma function, α represents the shape parameter, β represents the scaling parameter, γ represents the offset parameter, and w i represents the adaptive weight, Where η is a parameter that controls the smoothness of weight distribution, n is the number of features, represents the variance of the jth feature, represents the variance of the i-th feature;

[0071] Define a feature combination function C to enhance the feature function H i Combine features to obtain a multivariate feature sample set V k , where k represents the number of samples, the expression is,

[0072]

[0073] In the formula, is the kernel function, θ k Represents the parameters in the kernel function;

[0074] It should be noted that by adaptively weighting and enhancing individual features and using kernel function combinations to generate new low-dimensional feature spaces, not only the discrimination between features is improved, but also the dimensionality is effectively reduced, creating a richer multivariate feature sample set carrying key information for subsequent analysis, thereby enhancing the learning ability and generalization performance of the model.

[0075] S3. Based on the multivariate feature sample set, a device energy consumption prediction model is constructed to obtain the predicted energy consumption value.

[0076] First, for the multivariate feature sample set V k Each feature V kjApply Fourier transform, expressed as, F(V kj );

[0077] Where F represents Fourier transform, V kj represents the value of the jth feature, and j represents the total number of features;

[0078] Secondly, for F(V kj ) to perform feature weighting and combination, expressed as,

[0079]

[0080] Among them, w kj represents Fourier coefficients;

[0081] Then, perform the inverse Fourier transform F -1 , expressed as,

[0082]

[0083] Finally, based on the neural network model, the equipment energy consumption prediction model is constructed, and the expression is:

[0084]

[0085] In the formula, E pred Represents the predicted energy consumption value, F and F -1 Respectively represent Fourier transform and inverse Fourier transform, which are used to capture the periodic information in the features, N NN (W,V k ) represents the output of a neural network model, W represents the set of network parameters used to learn non-periodic complex relationships, and k is the number of samples;

[0086] It should be noted that the model combines Fourier transform to analyze periodic patterns in energy consumption data and neural networks to learn non-periodic complex associations. By combining multivariate feature vectors, the accuracy and generalization ability of equipment energy consumption prediction are effectively improved.

[0087] S4. Input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value.

[0088] Collect real-time data ti , where t represents the time point, and the real-time multivariate feature sample set is obtained through preprocessing and feature combination formula Real-time multivariate feature sample set Input the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value E pred，t ;

[0089] It should be noted that this step collects and processes workshop data in real time, and converts it into model input format using preprocessing and feature engineering methods. Subsequently, the predicted energy consumption value at the current time point can be calculated through the prediction model, thus realizing dynamic monitoring and efficient management of energy consumption.

[0090] S5. Calculate the deviation between the real-time predicted energy consumption value and the actual energy consumption to obtain an evaluation result.

[0091] The energy consumption value E will be predicted in real time pred，t The actual energy consumption value E at the same time step actual，t Compare and calculate the deviation value ΔE t :

[0092] ΔE t =E actual,t -E pred,t

[0093] Set the deviation threshold to A = 5%;

[0094] When ΔE t When ≤5%, it indicates that the equipment energy consumption prediction model is highly consistent with the actual energy consumption, and the equipment energy consumption prediction model is used;

[0095] When ΔE t When >5%, it indicates that the prediction deviates greatly from the actual situation. By adjusting the standard deviation σ of the kernel function, the adjustment parameters α, β, γ, and the adaptive weight w i , kernel function parameter θ k , optimize the equipment energy consumption prediction model until ΔE t ≤5%;

[0096] Furthermore, in the case of a large deviation between the predicted and actual energy consumption, in addition to adjusting the model parameters, it is necessary to deeply analyze abnormal events or external factors in the production process (such as sudden weather changes, equipment aging, etc.) and incorporate these factors into the model considerations;

[0097] It should be noted that this evaluation process compares the deviation between the predicted and actual energy consumption in real time, uses MSE to quantify the prediction accuracy, guides model optimization, and ensures that the model prediction closely matches the actual situation, thereby achieving high accuracy in energy consumption prediction and continuous iterative improvement of the model, providing a reliable prediction tool for energy management.

[0098] This embodiment also provides an evaluation system for power consumption of a factory workshop, including a data collection and preprocessing module, a feature engineering module, an energy consumption prediction model building module, a real-time energy consumption prediction module and an energy consumption evaluation module.

[0099] Specifically, the data acquisition and preprocessing module is used to obtain basic workshop data and preprocess the basic workshop data;

[0100] The feature engineering module is mainly used to combine the features of the pre-processed workshop basic data to generate a multivariate feature sample set;

[0101] The energy consumption prediction model building module is used to build a device energy consumption prediction model based on a multivariate feature sample set to obtain a predicted energy consumption value;

[0102] The real-time energy consumption prediction module is used to input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value;

[0103] The energy consumption evaluation module is used to calculate the deviation between the real-time predicted energy consumption value and the actual energy consumption to obtain the evaluation result.

[0104] This embodiment also provides a computer device, which is suitable for the method of evaluating the power energy consumption of a factory workshop, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method of evaluating the power energy consumption of a factory workshop as proposed in the above embodiment.

[0105] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WiFi, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0106] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for evaluating the power consumption of a factory workshop as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0107] In summary, the present invention evaluates the power consumption of factory workshops, uses multiple types of sensors to collect comprehensive data, and improves data quality through preprocessing; strengthens and combines features through innovative algorithms, reduces dimensions while retaining important information, and establishes a prediction model that combines Fourier transform and neural network to accurately predict energy consumption; real-time data is input into the model to calculate the deviation value, and the prediction accuracy is evaluated with the mean square error to guide energy consumption management and strategy adjustment.

[0108] Example 2

[0109] Referring to Table 1, which is the second embodiment of the present invention, in order to further verify the advancement of the present invention, experimental simulation data of the evaluation method of power energy consumption in a factory workshop are provided.

[0110] In order to verify the innovation and practicality of the present invention, a detailed experiment was designed. The experiment was conducted in a typical manufacturing plant workshop to accurately evaluate power consumption and optimize the prediction model. First, we installed power metering sensors, environmental sensors (including temperature and humidity), and equipment status monitoring sensors at key locations in the workshop. These sensors collected data once a minute to ensure the real-time and comprehensiveness of the data.

[0111] In the data preprocessing stage, we first process the original data d i Minimum and maximum scaling were performed to ensure that all data were in the same scale space. Subsequently, a Gaussian filter was applied, where the standard deviation σ was set to one tenth of the data volatility, effectively removing high-frequency noise in the data. The preprocessed data is recorded as

[0112] Next, the weighted feature enhancement function H defined by the formula i Enhance each filtered eigenvalue and calculate the adaptive weight w i To balance the importance of each feature. The feature combination function C further integrates the enhanced features, achieves dimensionality reduction through the kernel function K, and enhances the generalization ability of the model.

[0113] The equipment energy consumption prediction model constructed combines Fourier transform and neural network, and the model parameters are optimized through multiple rounds of iterations to minimize the prediction error. After the real-time data is processed through the same preprocessing process, it is input into the model to obtain the real-time predicted energy consumption value E. pred，t .

[0114] Finally, by comparing the predicted energy consumption value with the actual energy consumption value E actual，t , calculate the deviation value ΔE t , which is used as the basis for evaluating the performance of the model.

[0115] The details are shown in the following table:

[0116] Table 1 Experimental record table

[0117]

[0118] From the above experimental records, we can observe that at different times of the day, our prediction model can accurately predict the actual energy consumption. For example, in the time period from 08:00 to 16:00, the average deviation ΔE between the predicted energy consumption value and the actual energy consumption value is t At just 2kWh, this represents a significant improvement in accuracy relative to the industry standard of 5kWh deviation.

[0119] Especially at 16:00, although the actual energy consumption reached the peak of 120kWh in a day, the model predicted value was 119kWh, with a deviation of only 1kWh, which proved the prediction stability of the model in high energy consumption periods. In addition, by comparing the deviation values ​​of different time periods, it was found that by adjusting the parameters of the feature enhancement function (such as α, β, γ) and the adaptive weight w i , which can effectively reduce the prediction deviation, which reflects the effectiveness of the feature enhancement and kernel function parameter optimization strategy proposed in the invention.

[0120] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for evaluating power consumption of a factory workshop, characterized in that: include, Install various sensors to obtain basic workshop data and perform pre-processing; Based on the preprocessed workshop basic data, feature combination is performed to generate a multivariate feature sample set; Based on the multivariate feature sample set, the equipment energy consumption prediction model is constructed to obtain the predicted energy consumption value; Input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value; Calculate the deviation between the real-time predicted energy consumption value and the actual energy consumption to obtain the evaluation result; The basic data of the workshop include current, voltage, power factor, active power, reactive power, temperature, humidity, light intensity, air quality, vibration, speed, pressure, and flow; The pretreatment comprises the following steps: First, define the basic data of the workshop as d i , for d i Perform minimum and maximum scaling to obtain Then, Gaussian filtering is applied to remove noise. After filtering, we get Where σ is the standard deviation of the Gaussian kernel, k represents the position of the current data point, j is the index within the window, ∑ j Refers to all d in the window ij Perform summation; The step of combining features to generate a multivariate feature sample set includes the following steps: Define a weighted feature enhancement function H i , where i = 1, 2, ..., n, expressed as, In the formula, represents the eigenvalue after the i-th filter, Γ is the gamma function, α represents the shape parameter, β represents the scaling parameter, γ represents the offset parameter, and w i represents the adaptive weight, Where η is a parameter that controls the smoothness of weight distribution, n is the number of features, represents the variance of the jth feature, represents the variance of the i-th feature; Define a feature combination function C to enhance the feature function H i Combine features to obtain a multivariate feature sample set V k , where k represents the number of samples, expressed as, In the formula, is the kernel function, θ k Represents the parameters in the kernel function; The construction of the equipment energy consumption prediction model to obtain the predicted energy consumption value includes the following steps: Based on the multivariate feature sample set V k , build the equipment energy consumption prediction model, expressed as, In the formula, E pred Represents the predicted energy consumption value, V k =[V k1 , V k2 , ..., V kj ] represents the multivariate feature vector of the kth sample, where V kj is the value of the jth feature, j represents the total number of features, F and F -1 denote Fourier transform and inverse Fourier transform respectively, w kj represents the Fourier coefficient, N NN (W,V k ) represents the output of the neural network model, W is the set of network parameters, and k is the number of samples.

2. The method for evaluating power consumption of a factory workshop according to claim 1, characterized in that: The various sensors include power measurement sensors, environmental sensors, and equipment status monitoring sensors.

3. The method for evaluating power consumption of a factory workshop as claimed in claim 2, characterized in that: Input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value including: Collect real-time data ti , where t represents the time point, and the real-time multivariate feature sample set is obtained through preprocessing and feature combination formula Real-time multivariate feature sample set Input the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value E pred，t .

4. The method for evaluating power consumption of a factory workshop as claimed in claim 3, characterized in that: The deviation between the real-time predicted energy consumption and the actual energy consumption is calculated, and the evaluation results include: The energy consumption value E will be predicted in real time pred，t The actual energy consumption value E at the same time step actual，t Compare and calculate the deviation value ΔE t , expressed as, ΔE t =E actual，t -E pred，t ; Set the deviation threshold to A; When ΔE t When ≤A, it means that the predicted energy consumption value is highly consistent with the actual energy consumption, indicating that the equipment energy consumption prediction model is normal; When ΔE t When >A, it means that the prediction deviates greatly from the actual situation. By adjusting the standard deviation σ of the kernel function, the adjustment parameters α, β, γ, and the adaptive weight w i , kernel function parameter θ k , optimize the equipment energy consumption prediction model until ΔE t ≤A.

5. A factory workshop power consumption evaluation system, based on the factory workshop power consumption evaluation method according to any one of claims 1 to 4, characterized in that: It includes data collection and preprocessing module, feature engineering module, energy consumption prediction model building module, real-time energy consumption prediction module and energy consumption evaluation module. Specifically, the data acquisition and preprocessing module is used to obtain basic workshop data and preprocess the basic workshop data; The feature engineering module is mainly used to combine the features of the pre-processed workshop basic data to generate a multivariate feature sample set; The energy consumption prediction model building module is used to build a device energy consumption prediction model based on a multivariate feature sample set to obtain a predicted energy consumption value; The real-time energy consumption prediction module is used to input the real-time multivariate feature sample set into the equipment energy consumption prediction model to obtain the real-time predicted energy consumption value; The energy consumption evaluation module is used to calculate the deviation between the real-time predicted energy consumption value and the actual energy consumption to obtain the evaluation result.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for evaluating the power consumption of a factory workshop according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for evaluating the power consumption of a factory workshop according to any one of claims 1 to 4 are implemented.

Images